They have offices all over the country at college towns for this very reason (see Google Ann Arbor)

https://www.google.com/about/company/facts/locations/

while there's a number of engineers at Google A2, it would constitute less than 10% of the headcount at that office (this is guess based on the LinkedIn research I've done. If someone has a more precise number, would very much like to hear it)

I've been there for interviews and tours - it is primarily ad sales and account managers.

This information may be out of date, its probably been three or four years since i was there.

I agree, open up site offices near universities, major labs, single tech company towns, etc. There's a huge number of talented people looking for work in those areas and it saves on everything from real estate expense to relocation expenses.

But then of course, big companies will do it wrong. I'm reminded of companies moving to places like the D.C. area, where all their clients are in Virginia or Maryland, but go ahead and set up shop in D.C., where nobody from their client locations likes to go. Even better they'll pick one of the few spots in the city without metro access and make it even harder on themselves to do business.

Companies almost need relocation experts for setting up these kinds of remote sites.

As a real estate developer, I would love to see some more discussion on this thread. What would it take to get a major tech company to relocate an office or two from Bay Area to much cheaper regions. Obvious items would be network effects, primarily access to skilled labor and capital. What else? Cheap, frequent flights to the mothership? Has Google fiber and the likes had any real impact on luring tech workers out? If anyone knows of existing discussion or papers on the topic, id love to read it. I'm in Memphis TN, and cities in the area have low cost of living, some dense urban areas, etc. Oh, and 30-40 story class A office towers that are partially filled.

People have to want to live there, otherwise you're very limited in your recruiting pool to people either already in that region, or possibly people from that region.

As an example, I work in Oslo, and we have offices in Trondheim and Tromso as well due to the universities in those cities. Recruiting people to work in Oslo is quite easy, but almost nobody is willing to move to Trondheim or Tromso unless they were looking to move back there anyway.

Opening office is not good enough.

The center of gravity also should move. Example, if all CXOs move then there is a chance that office will prosper and people from other parts of country would also move there even if they had no plans earlier.

Quite often I have seen that satellite office don't prosper because the center of action is still at HQ, that translates to freedom of moving between teams, opportunity to work on problems that really matter for the company, etc. Satellite office attracts only those folks who have reasons to be in those cities (usually for personal reasons, very seldom for professional ones).

Only way it can be really effective is if a company moves lock, stock and barrel to a new location - the way Boeing did in early 2000 (moved its HQ from Seattle to Chicago).

Google already has offices all around the world. They are already doing this.

But aren't a lot of those offices either 1. much smaller than MV or 2. not engineering offices?

This happened in Pittsburgh. Not quite a college town, but many Google Pittsburgh employees are straight out of Carnegie Mellon.

It's definitely a recruiting strategy. Heck, until fairly recently, Google Pittsburgh was on the CMU campus.

Apple, Intel, Disney, and a bunch of others still are.

http://www.cmu.edu/corporate/partnerships/cic/tenants.html

Google has an engineering office here in Pittsburgh right down the road from CMU. So do a lot of other tech companies. A huge part of that is the sheer number of colleges in the area, low cost of living, and apparently the booming food scene with lots of creative chefs due to low rents [0]. Not sure if I'm wiling to buy that last one as a tangible factor in economic growth, but it certainly tastes good in the interim.

0. http://www.nytimes.com/2016/03/16/dining/pittsburgh-restaura...

Might you be able to test this out for yourself? Is your job location the main reason you don't live in your university town? For me it's not.

Right, you mean like Google Waterloo

It can be hard to find people who actually want to live there. It's one thing to have a stream of new interns and even fresh college grads, but what do you do when they constantly want to move somewhere else after a year or two?

It'll be more and more difficult to convince people to move to Mtn View, though.

It's a delicate balance.

Certainly cost of living in the area is becoming a major issue for companies. But it's the startups who are already being priced out of the area. Companies like Google and Apple can afford to offer higher wages (even if they don't necessarily want to.)

But before you see these companies trying to start a major hub in a university town, there are other large cities that would be more palatable: Chicago, Pittsburgh, Boston, and Portland are places I would expect to be considered. Even Raleigh/Durham and Charlotte since there's already a fair bit of momentum there, although I suspect them to be a more difficult sell for the same reasons I think Atlanta will have trouble attracting talent.

Atlanta would be somewhat attractive, except that Georgia has just passed their "Religious Freedom" law (i.e. the "We Support Discrimination" law). Tech companies tend to be more liberal and don't like limiting their applicant pool, and setting up shop there would certainly do so.

Mountain View and the Bay Area, maybe. But they also have offices in Austin, Seattle, NYC, etc.

They do have those, many are from acquisitions like Chapel Hill.

Why stop even there - why not just build your own town and your own university to train people the Google Way - if Trump can have a university, why not Google?