Speaking from the perspective of working remotely for over 5 years with multiple teams (all of differing compositions): salaried + contractors, US (all three ti

Speaking from the perspective of working remotely for over 5 years with multiple teams (all of differing compositions): salaried + contractors, US (all three timezones) + South America + Europe (Eastern and Western) + Asia, junior + senior + staff engineers, etc.; and being involved in plenty of interview loops for remote engineering positions. You're missing two key reasons why salaries won't normalize to developing country level pay: 1) Timezones are painful. +3h west to east coast US is usually tolerable if not slightly inconvenient at times, but not too bad. +5h and above starts to strain team dynamics. East US to London/Ireland caused more than a couple great US engineers to seek greener pastures within other teams at the company; a few teams had difficulty staffing majority European teams with US talent. A 50\50 split was usually more tolerable. Any time difference more than that within a team is insanity. Feedback loops become too large, there's no cohesion, and a host of other issues. 2) 10M+ engineers at the senior level don't exist. You want engineers that can autonomously churn through "we have a problem with no identified solution" situations. These engineers earn 2x to 3x their cost through technical leverage. At the senior+ level, talent supply is lower than demand. And if you want to train up an employee from junior to senior, good luck doing that with a timezone difference > +3h. As long as there are tech companies printing gobs of money while demanding more talent than is available for engineers to produce $$$ for $, then salaries won't normalize to developing country pay.