Analogy: Imagine that instead of TCP/IP protocol running the Internet today if we had instead 'Aussie Network Protocol' for Australia, 'Canada Network Protocol' for Canada and 'German Network Protocol' for Germany, etc, etc. Communication between these different protocols is possible through conversion processing at gateways. This causes Inter-Country communication to become massively bottlenecked. This creates a misalignment of incentives whereby participants are enticed to seek domestic content alternatives of a lesser quality. Massive amounts of capital are then spent on increasing the conversion throughput (Which would be unnecessary if working off the one protocol) and less eyeballs are looking at each different protocol for security holes (Because they are divided over the number of different protocols). International companies wishing to setup services incapable of being provided through the gateway throughput limitations now have to employ engineers in each country to rebuild the same service.
This does not mean you can't have currencies / layers on top of Bitcoin (Pegged). Much like how Australia has the AU country code top level domain, and Canada has the CA country code top level domain, but all of the internet services are running on TCP/IP.
Your strawman argument is therefore quite transparent. Massive mis-allocation of incentives and resources due to conversion between protocols, does not equal 'International communication / investments is a huge negative'.
