I still think what I said is true basically by definition, but yeah fair enough it does beg the question why companies don't hire exclusively in cheap areas now

I still think what I said is true basically by definition, but yeah fair enough it does beg the question why companies don't hire exclusively in cheap areas now, driving down the market rates in expensive cities. I think supply and demand is still a big part of the answer. By square mileage the potential locations where employees can live has grown by 200x by opening up to remote work anywhere in the US (napkin math, US is 3.8 million sq miles, bay area and tri-state area combined are ~20k sq miles). But obviously the number of available engineers hasn't expanded by nearly that much, engineers are much more concentrated in the areas where most of the jobs have been concentrated up until now. I.e. companies still can't afford to limit their supply to remote only, just as they increasingly can't afford to limit supply to bay area only. Then there's also the perhaps more controversial issue of skill & experience. The most in-demand engineers are the people with experience at Google, FB, etc. and these people are concentrated in the places where these companies are. Ambitious people also have self-selected to live in these hot markets for years. You could argue that these proxy factors shouldn't matter and companies should be measuring skill directly, but as an industry we still haven't cracked that problem. So on average, a remote engineer in NYC probably is more valuable to most companies than one in Birmingham, Al. Like it or not, there's still a lot inertia to overcome here (and maybe even a little bit of validity to these patterns).