I didn't ask, but my guess would be that it was net of fees but not net of any taxes the traders would have paid. The brokerage house would have no way to know the latter even though it eats into the net by a substantial amount. But actually, the more important thing is that it almost certainly could not have considered survivorship bias. This comes from traders who close their accounts during the year, most probably because they lost money. It would be very hard to identify those accounts and include their losses in the 1.5% net, so the net is probably based only on accounts that existed for the full year. My guess is that if survivorship bias was properly considered, the net for the year would be negative.