Lucas (1988) hypothesised that human capital externalities explain persistent productivity growth and become manifest via interactions between workplace colleagues. Consistent with the first part of this hypothesis, Fox and Milbourne (2006) concluded that an increase in the average level of human capital in Australian economics departments raised the research productivity of departmental members. This paper tests the robustness of this finding by using a direct, rather than a proxy, measure of human capital and confirms the existence of human capital externalities within Australian economics departments. But we extend the analysis in two important dimensions. Firstly, we investigate the second part of Lucas' hypothesis by testing whether the externality becomes manifest via co-authoring. We find no evidence that this type of interaction is associated with higher research productivity, especially for higher quality outputs. Secondly, we control for the likely endogeneity of one's peer group via instrumental variables estimation. In this case, we find that the peer group effect disappears completely for the highest quality outputs but remains for research output more broadly defined.