One of the commonly expressed intents for buying indices is lazy diversification. By having vastly unevenly-weighted indices, one fails to accomplish that aim.

One of the commonly expressed intents for buying indices is lazy diversification. By having vastly unevenly-weighted indices, one fails to accomplish that aim. Assuming diversification is the goal, your comment would only be true (that 1+1 and 2 are equivalent) if the constituent parts of hyper-scale companies operated independently. In reality, I'd say that rarely happens. At best, you get something like Facebook (there can only be so many true chiefs). At worst, you get something like GE (cannibalizing viable businesses to feed others). And all of this is to say nothing of that fact that, historically, Amazons don't become 2x Amazons with the frequencies that {mid-cap} become {large cap}, simply by virtue of total addressable market maximums in mature business areas.