I found a couple of papers making a claim that is likely to be controversial: a researcher could be anywhere in the world and still get as much research done. This would be new effect from around 1990 (birth of the web?).
I’ll just cite the juicy parts and let you decide.
The disappearance of positive university fixed effects suggests that, as far as research productivity is concerned, elite universities are indeed losing their competitive edge. Because of the reduced importance of personal interactions, one traditional advantage of elite universities – to act as a focal point attracting the smartest faculty – is at risk.
Reference: E. Han Kim, Adair Morse, Luigi Zingales, Are Elite Universities Losing Their Competitive Edge?, NBER Working Paper No. W12245, May 2006.
(…), we find that the spillover effects associated with location [near a research superstar] have declined markedly over time. (…) These results are puzzling. One would expect the diffusion of the internet to diminish the role of geographic distance as a factor influencing the extent of spillovers, but not to the extent of making physical proximity irrelevant, or a hindrance.
Reference: Pierre Azoulay and Joshua Graff Zivin, Peer Effects in the Workplace: Evidence from Professional Transitions for the Superstars of Medicine, working paper (2005)
(Original source: Stephen Downes)