This paper examines whether public R&D subsidies constitute a substitute or complement for private-financed R&D. The empirical analysis is based on a panel data of 223 Japanese high-technology start-ups. Our evidence is consistent with the complement hypothesis, i.e., that publicly-funded R&D does promote private R&D. The complement effects are stronger for more mature firms. This is because such firms, in the growth phase, might have greater demands for R&D funds.