Micro-based and macro-based approaches have been used to assess the effects of health on economic growth. Micro-based approaches aggregate the return on individual health from Mincerian wage regressions to derive the macroeconomic effects of population health. Macro-based approaches estimate a generalized aggregate production function that decomposes output into its components. The microbased approach tends to find smaller effects than the macro-based approach, thus presenting a micromacro puzzle regarding the economic return on health. We reconcile these two strands of literature by showing that the point estimate of the macroeconomic effect of health is quantitatively close to that found by aggregating the microeconomic effects, controlling for potential spillovers of population health at the aggregate level. Our results justify using the micro-based approach to estimate the direct economic benefits of health interventions.