The current study focuses on the cumulative effect on children's early learning outcomes of the quality of teacher-child interactions over multiple years during early childhood. Using propensity score matching to minimize selection bias, we compared the academic outcomes of children who experienced consistently high-quality or consistently low-quality teacher-child interactions, as defined by the Classroom Assessment Scoring System, during prekindergarten and kindergarten. Research Findings: Results indicated statistically significant differences in directly assessed language and literacy skills between children who experienced consistently high- and consistently low-quality instructional support. This was true after just 1 year, and gains were even greater for children who experienced high levels of instructional support over 2 years. Emotional support and classroom organization were not associated with children's academic learning. Practice or Policy: Although children experience cumulative benefits in response to high-quality instructional supports over 2 years, relatively few children have access to such opportunities. We discuss these results in light of limited consistency in children's experiences over the early school years and the presence of selection bias.