> But that's not true, because suppose if 99.9% of investors are buying ETFs and only 0.1% are actively managing, there won't be enough funds to bring all these

> But that's not true, because suppose if 99.9% of investors are buying ETFs and only 0.1% are actively managing, there won't be enough funds to bring all these panic dislocated stocks back up for many, many years. The price change of a stock doesn't always depend on how often a stock is bought. For example, when earnings hit, a stock can easily move 10%. That's not because there was a certain amount of people buying at every price between what it was and +10%. This can happen because earnings are better, and many people base price of an earnings multiple. When new knowledge is put into the system, there's a new price. If everyone who is trading a stock agrees on a price, that is the price. If people don't agree, it is generally the highest price someone is willing to sell it for (and find a buyer), or the lowest someone is willing to buy it for (and find a seller). If that disconnects a lot, you get a high bid/ask spread, and possibly no shares trade (lack of liquidity). It might only take a few shares or be a very short time to move a stock a lot.