South Africa's basic education system is dysfunctional. It scores last or close to last in a myriad of metrics and delivers learners with some of the worst literacy and numeracy competencies worldwide. A bimodal distribution in the results exists when learners from the richest socioeconomic quintile are performing adequately well, while learners from the poorest quintiles are failing. This paper presents a system dynamics simulation model to describe the causal linkages between improved early childhood and pre-school learning practices on the education system as a whole. The paper investigates the difference in performance between rich and poor communities. Three interventions explore the research question of whether it is the number of enrolments into early childhood development programs that increases a cohort's school readiness, or rather the quality of the early childhood development programs into which they were enrolled. The results answer the research question for the Western Cape province by showing that increasing the quality of the formal ECD programs leads to a greater percentage of school-ready five year olds than increasing the percentage of enrolled children, but that decreasing community poverty leads to better results than either intervention. The results show the simulation model to be a powerful tool to assist with policy setting and intervention testing for any other province or country by simply changing the input data and calibration.