macrocephalopod

the same point that the *dollar* volume of buying is larger as if saying it louder will make it more true (hint: if dollar volume is what matters then why doesn't Apple see a massive price impact every time SPY gets a new inflow?)

So it's not mechanical buying and it's not flows. What do I want to say then? Essentially that a high passive ownership of a stock *lowers the threshold* for market structure breakdowns which can cause large price swings in individual names, e.g. what happened to $GME

I can build a simple dumb model for this. Say that a stock has F shares in its float and there are S shares being shorted, which means that there are F+S shares held long. Assume that p% of the free float is held my passive vehicles who, to a first approximation, will never sell.

That means there are (1-p)F+S possible sellers. Now assume some fraction d of the possible sellers have diamond hands and will also never sell. How large does d need to be before a really epic short squeeze becomes possible?

Roughly this would happen when there are more shares sold short than willing sellers i.e. when S > (1-d)[(1-p)F+S] which rearranges to d > (1-p)/(1-p+S/F)

longs who need to coordinate to never sell becomes surprisingly small, especially for highly shorted stocks - something like this happened with $GME where even though there were willing sellers (e.g. Fidelity who was one of the biggest holders and sold virtually everything)

a relatively small (25%?) but committed proportion of longs who agreed to not sell was sufficient to squeeze the price to stupid levels, at least until sufficiently many shorts had been covered.