The classic Beckerian framework suggests that high-educated parents, who spend less time on home production and leisure but more on work, should spend less time with their children, since their non-market activities have higher opportunity costs. However, previous research finds the opposite, leading to a “parental time-education gradient” puzzle. I develop a model of intra-household time and resource allocation that provides a framework for disentangling the four relevant mechanisms for the puzzle suggested by Guryan, Hurst and Kearney (2008): high-educated parents’ differing (1) time efficiency in child human capital production, (2) preference for child human capital, (3) substitutability between time and market goods in child human capital production, and (4) wage capturing the opportunity cost of non-market activities. I assess the model’s predictions by exploiting an exogenous decrease in the daycare price in Québec. First, I find that parents increase time devoted to their children at the expense of home production while consuming more home production market goods (eating out, hiring domestic help) and child market goods (daycare, games, toys); this suggests complementarity between time and market goods in child human capital production, and substitutability in home production. Second, I find the time reallocation to be larger for high-educated parents; this, through the lens of the model, suggests that mechanisms (1) and (2) dominate over (3) and (4). I use the model to estimate structural parameters, which buttress the pivotal role of complementarity between time and market goods in child skill formation, and also indicate high-educated parents’ (1) stronger time efficiency in child skill formation, (2) stronger preference for child human capital, (3) and stronger complementarity between time and market goods in child skill formation. I use the structural estimates to assess how universal daycare shapes the early childhood skill gap. The counterfactuals reveal that after a daycare price decrease, the relative importance of the time efficiency, wage, and preference channels is approximately 1:2:4 in explaining the skill gap. Furthermore, a lower bound of the relative importance of the time efficiency and preference channels is 2:1 in explaining high-educated parents’ larger child time increase. These results substantiate that high-educated parents’ time efficiency advantage in child skill formation, besides their larger (wage-driven) parental investments and stronger preference for human capital, is a crucial mechanism behind widening early childhood skill gaps.