Using administrative data from firms in Australia that conduct research and development (R&D), we examine how R&D activity of other firms and public institutions affect a firm's own R&D expenditure. We distinguish between the impact of peers, suppliers and clients. We examine whether geographical proximity and industrial clustering affect R&D spillovers. Overall, we detect positive effects on R&D expenditure from spillovers from peers and clients to firms that are nearby; within 25 or 50 km. R&D expenditure by academia, unlike by government bodies, has a positive influence on a firm's own R&D expenditure within state boundaries. We fail to find any significant role for industrial clusters in augmenting spillover effects.