[js8]

When things are produced locally, you get closer to consumer and feedback loop is shorter. (That's why Shenzen is so great as a HW startup hub - if you can get a part in minutes compared to days, prototyping is faster.) You can understand the practical problems with your products better if you can more interact with customers.

[jsbg]

Isn't this priced into the product? I don't see how this contradicts comparative advantage.

[js8]

No, see https://news.ycombinator.com/item?id=44866613. If the theory of comparative advantage is to give any useful advice, it cannot just rely on people doing "the right thing". That will quickly make it into an unfalsifiable tautology.

In this case, you also forget that the pricing signals have a time horizon, which is bounded by the expected time of survival of the private company being non-competitive. But government (society) can take a longer term view than a company that is influenced by pricing signals, See the LG example I gave elsewhere in this thread.

Another reason is incomplete information. Free market works in theory because you assume complete information; in the real world, this is not the case.

[jsbg]

> the theory of comparative advantage

It's not a theory. Firms trade internationally because it's cheaper, not because they "believe" in free markets.

> But government (society) can take a longer term view than a company that is influenced by pricing signals

Private markets routinely make multi-decade bets; entire industries (mining, energy, infrastructure, biotech) work on 20–100 year horizons. Risk and time are already priced in.

> Free market works in theory because you assume complete information

Free markets don’t "assume" anything; they function through decentralized price signals. And even with imperfect information, dispersed actors still aggregate and react to knowledge far better than any central planner can.