macrocephalopod

This is not true! A really basic cap-weighted index fund mechanic is that they automatically track the index as prices move, with no trading required. This is literally the reason that they are called *passive* funds.

Interesting, could you elaborate a bit? So if one component rises and becomes a larger part of the index, are you saying, the fund doesn't have to buy more, bc it already owned that component, and when it rises, the funds' exposure to it rises too, so no need 2 adjust?

Yes, this is exactly correct!

I honestly believed that this was so widely known that it didn’t need explaining but seeing three apparently smart people get it wrong has convinced me otherwise.

Still, every new round of flows will be biased toward the components that have risen the most since the last round of flows, right? If true than there is some momentum effect going on, flows will favor those that have risen the most in betw last round, could accel trends perhaps?

On paper if you go for full replication of an index you might have to buy more of the "big" stocks, but that's supposing there are significant net positive flows. That could have an impact indeed but I doubt ETFs have such large daily inflows (I may be very wrong though)

The bigger impact from large net inflows to a cap-weighted ETF is that it creates buying in all index constituents which pushes all prices up in tandem. Much less effect on relative prices precisely because they are cap-weighted, and market cap correlates strongly with liquidity

Story is very different for non-cap-weighted funds (e.g. ARK ETFs) where inflows have a much bigger price impact on the names that the fund has biggest positions in (where “biggest” means “largest proportion of market cap”)