Not gonna lie,
this article influences this content.
First, let me say that I don't think indexes are socialism so long as people invest in them by their own free will. This is similarly false:
Indexation could work if it remained a satellite strategy, with say 10% of the money being managed through indexation, the rest being managed by active money managers. As such, it would be a parasitic strategy. Indexers would benefit from the price discovery work done by others without paying the costs associated with the process.
Since a money manager improves price discovery when alpha is obtained (i.e, buy low sell high means that you push the price up through buying when the price is low, and push the price down by selling when the price is high), the only investors that are helping with price discovery are those that outperform the index, an outperformance which, of course, they are getting rewarded for, while those investing directly in the index are not.
But, the article does bring up an interesting point. Namely,
Active money management is essentially a ‘mean reversion’ strategy. That’s not so for indexation. In the indexation process, there is no attempt at price discovery. The only thing that matters is the relative size of the asset: the bigger the market capitalization, the more an investor should own. This means if the price of a large asset goes up more than the market as a whole, indexers have to buy even more of it.
When you're saying it that way, you realize that it really is quite weird, and opens the door for all kinds of market manipulation.
That seems like an incredibly odd system.
Why not index by earnings, instead of by market cap? That way your index actually buys a larger % of companies with a low PE, than companies with a high PE, among companies with the same EPS?