As a counterpoint, in the leadup and when Amazon announced their HQ2 in Northern Virginia, one thing they were very clear about is that no teams and no employees except some execs were going to actually get reassigned to the new location. All staffing at the new location would be via new hires at the new location. They knew full well that if they tried to move a team, most of the members of the team would transfer to a team that wasn't relocating, and that if they tried to prevent that the employees would just quit.
Relevant story from Peopleware (Tom DeMarco and Tim Lister):
One thing Townsend seems to have missed entirely is the presence of women in the workforce. The typical person being moved today is part of a two-career family. The other half of that equation is probably not being moved, so the corporate move comes down hard on the couple’s relationship at a very delicate point. It brings intolerable stress to bear on the accommodation they’re both striving to achieve to allow two full-fledged careers. That’s hitting below the belt. Modern couples won’t put up with it and they won’t forgive it. The company move might have been possible in the 1950s and 1960s. Today it is folly.
Even in the sixties, organizational moves didn’t make much sense. A case in point is the decision by AT&T’s Bell Laboratories to move the 600-person ESS project from New Jersey to Illinois in 1966.
[...]
Years after the ESS cutover, I arranged to interview Ray Ketehledge, who had run the project. I was writing some essays on management of large efforts, and ESS certainly qualified. I asked him what he saw as his main successes and failures as boss. "Forget the successes," he said. “The failure was that move. You can't believe what it cost us in turnover." He went on to give some figures. The immediately calculable cost of the move was the number of people who quit before relocation day. Expressed as a percentage of those moved, this initial turnover was greater than the French losses in the trenches of World War I.
[...]
And that accounts only for the initial loss. In the case of Bell Labs, there was another large exodus starting about a year after the move. These were the people who had honestly tried to go along with the company. They moved and, when they didn’t like the new location, they moved again.
Counter-counterpoint, Toyota consolidated several offices in CA/KY/NY to one HQ to Plano, TX and brought about 2800 out of 4000 employees. It was a big disrupter to local housing prices, with massive swaths of land experiencing the maximum-by-law 10%-yearly property tax increases.
I don't know if there's any easy generalizations to suss out. Probably every case will be very specific to the work being done, the environment it's done in, and the people it's done by.
