In this paper, we examine the income gradient in child mental health using longitudinal data from a large, national cohort of Australian children. We contribute to the body of existing literature by: (i) investigating whether and to what extent a child’s mental health levels and their relationship to income vary when a child’s mental health is assessed by the child’s parent, the child’s teacher and the child her/himself; (ii) exploring whether the reporting differences in a child’s mental health is associated systematically with household income; and (iii) examining the child mental health gradient and the evolution of this gradient by the child’s age. We found that a child’s mental health and the income gradient vary depending on who assesses the child’s mental health (the gradient was the largest when assessed by parents and the smallest when assessed by the child). Furthermore, the magnitude of the effect of mental health and income gradient faded when we controlled for some important variables, such as maternal health.