Los Alamos National Lab is a 12,700 person [1] government research facility in a town of 13,200 population [2]. Median household income $98k, highest millionaire concentration of any US city, highest percentage of people with doctorate degrees in the nation. Nearest IKEA store? 500 miles away.
A government lab can get scientists to relocate their families to the middle of nowhere. All it takes is a budget of $3.92 billion or $300k per job.
By all accounts it's very hard for grad students and postdocs at LANL (much less locals) to find affordable short-term housing near LANL due to (1) the very small amount of housing available and (2) enough people leaving the lab but continuing to live there while taking up remote work in tech. This is despite LANL paying relatively well (as you noted) to try to keep talent from moving into other industries (tech, finance).
Not to say that government investment can't make people move, but it creates problems too, not least for locals.
Source: mainly anecdotal, from talking to people at LANL at various career stages on a recent visit.
I haven't been there in a while, but last I recall, they sort of ran out of land. It's very hilly, the airport basically ends at a cliff. A lot of the land is government controlled and fenced off to provide buffer for secure areas.
A lot of land around that is national park/BLM land, and not accessible. So you've got like, white rock, espanola or Santa Fe to commute in from. I did know a guy who had a light plane and would commute from the east side of Albuquerque to Los alamos by plane.
So, sort of like local zoning, but more like DC. an act of congress to free up some of that land, not as straightforward as visiting some city council meetings.
Taking a look at some randomly selected roads in Google Street View [1], it looks like it's been built at extremely low density, by most standards.
I mean, if you've building a single-storey home, then putting a double garage on the side, then a gap between properties two cars wide, plus a driveway with parking for two cars, then on-road parking for 3 more cars? Of course you're going to run out of land fast.
Los Alamos is on a mesa, with cliffs on one side and Jemez mountains on the other.
There were some local controversies related to zoning when I was there (namely, commercial real estate), but the issue is much broader and long-standing.
I used to work for a PI who, during his internship days in the ~1980s, lived in a tent in the mountains the whole summer (he was a delightfully quirky person though).
Los Alamos got kickstarted in WWII when the government literally got to tell physicists where to go and it became a major hub of such research.
That might work if the NIH wanted to spend 50% of it’s budget in one location, but their proposal was to spread across not just one second tier city but dozens of such locations.
The scientists apparently had some wiggle room: An offer was made to the Princeton team to be redeployed there. "Like a bunch of professional soldiers," Wilson later recalled, "we signed up, en masse, to go to Los Alamos.”
But the subtext was there was a draft going on and most of them would be eligible unless they were working on such a critical project.
No, that's the budget for everything - salaries, equipment, grants given to outside bodies, and so on.
I mention it merely because, although LANL demonstrates the state can create pockets of prosperity in poorer places by moving science jobs, we should not imagine that it's particularly cheap.
The LANL example also doesn't show economic success spreading out from the lab, in the way Silicon Valley is sometimes seen as spreading out from defence research and Stanford University. Of course, LANL is deliberately geographically isolated, and you wouldn't expect startup spin-outs from a top secret nuclear research lab.
Not to say that government investment can't make people move, but it creates problems too, not least for locals.
Source: mainly anecdotal, from talking to people at LANL at various career stages on a recent visit.