Returns to Scale in Active and Passive Management

We revisit the nature of returns to scale beginning with Pástor, Stambaugh, and Taylor (2015). We find that the documented negative relation between scale (at both the fund and industry levels) and return performance is an artifact of extreme observations that comprise less than 0.05% of the sample. A manual examination against outside sources shows that the origin of these observations are data errors that occurred during the dot-com crash. We then confirm constant returns to scale by revisiting Zhu (2017) where we find outliers driving diseconomies of scale results. We reconcile non-negative returns to scale with Berk and Green (2004) using index funds to proxy for the passively managed component of actively managed funds and find an inverse relation between non-expense-ratio operating costs and fund size. We argue this decline offsets diseconomies of scale in active management