> It's extremely difficult if not impossible to completely corner some market

Are we forgetting how Microsoft absolutely dominated the PC market in the 90s and early 2000s? How today, there's basically two mobile operating systems that can determine much of today's mobile experience?

I agree that IP laws serve to entrench monopolies (although they also have the effect of encouraging development (e.g. of expensive drugs) in the first place - so it's a double edged sword). But so do a lot of other, much more "natural" mechanisms such as network effects, brand recognition, etc.

I feel like the libertarian, "pure capitalism" world view only makes sense if you think that consumers are omniscient and perfectly rational, but we know fully well how neither of these things are true.

I used to work for a company that would just buy up a bunch of smaller companies and then drive them into the ground. The smaller companies had the better tech, but the large company had a user base, a brand and money.

The point about operating systems is that these are new technologies. Free markets require time to resolve. It's natural that an inventor (especially with IP protections) has an market advantage for a period of time. It's important, that at longer time scales, the market evolves towards decentralization.

It's true that there are a lot of natural centralizing forces, such as network effects. Not sure how to resolve those.

However, getting back to the original point, no matter how you think about it, capital efficiency requires that 'a few' manage the capital, because naturally there are 'a few' who manage capital better than the rest. The problem is selecting those few. I think it's a misconception to argue that capital shouldn't centralize in principle, or that it's a morally bad thing.

Also, efficiency in a free market is really about fulfilling the needs of people according to their free will in the most efficient way possible. There's no forcing function to produce more if people don't want to consume more. In Soviet Union, they were very efficient at producing some stuff, but they made stuff that no one needed (they made lots of cheap shoes in big factories, which no one wanted). I.e. there's a difference in being efficient in isolation, and efficient at fulfilling needs of the market.

First, I don't think that "capital efficiency" is something that we absolutely need to optimise for regardless of any other concerns, and even if I would concede that, there's an argument to be made that concentrating capital in the hands of more people leads to more diverse markets that cater to more, and more different, people's needs. The thesis that a few number of people are just naturally more efficient at everything seems questionable to me.

It's true that there's a lot of specialization required and other reasons why capital doesn't always scale. That's basically what limits everything concentrating to just one guy.