Though the positive income gradient of child health is well documented in developed countries, evidence from developing countries is rare. Few studies attempt to identify a causal link between family income and child health. Utilizing unique longitudinal data from the China Health and Nutrition Survey, we have found a positive, age-enhancing income gradient of child health, measured by height-for-age z scores. The gradient is robust to alternative specifications and a comprehensive set of controls. Using the fact that the rural tax reform implemented since 2000 created an exogenous variation in family income across regions and over time, we explore a causal explanation for the income gradient, and find that it has a very strong independent causal effect on child health.