I think the points about challenges or switching and the likely future of heterogeneous office/remote balances are all strong.
Where this lost me was on this claim about salaries:
> I suspect the multipliers companies use to determine wage per city will even out to the efficient frontier of wage to cost of living as employees allocate themselves to the optimal locations.
I think this is a temporary situation until there is a critical mass of remote roles/companies. If I live in a lower COL region but looking at remote roles I'm not competing with people near me, and the employers are not competing with employers near me. I'm competing with people who can do the work I do, and the employer is competing with all other remote-capable companies. I think the only shape of this that might work is treating the target time zone as a single market. Assuming comparable skills, a worker in Boston, Atlanta, NYC, Toronto, Nassau or Bogota shouldn't have wide compensation differences.
It's possible that this leads to a net reduction in salary for many, as big companies "figuring out" remote might make formerly-painful offshoring more workable and attractively inexpensive.