We estimate the causal impact of publicly funded R&D on private-sector productivity growth.
Identification is based on a novel external instrument derived from a narrative classification of all significant postwar changes to federal R&D appropriations. Using long-horizon local projections and our narrative instrument, we find that higher government R&D spending predicts persistent increases in private-sector productivity growth along with other measures of innovation. We estimate that the production function elasticity of government R&D capital is 0.12, implying that the returns to nondefense government R&D averaged roughly 180-211% over 1947-2022. Our estimates suggest that public R&D and public infrastructure investment each accounted for roughly 20-25% of business-sector productivity growth since 1947.