most EAs should probably not be living in high cost-of-living (HCOL) areas most of the time

By matthes @ 2026-07-02T19:30 (+21)

If you work a job that can mostly be done from anywhere^[1], you should ask yourself if you are spending too much money on housing. 

I felt prompted to write this post by the recent increased push for EAs to move to the Bay. Multiple people have urged me personally to relocate. Most of them want me to work on AI safety, but the move-to-the-Bay meme is now also strong in animal welfare. I remain unconvinced, although I am always open to visiting if there is something specific for me to do there.

the core case

Significant additional spare money isn’t just great for donating, but it can also make it easier to have an impactful career, as you can afford to: 

• take more risks (e.g. start a new org that might fail)
• spend more time on your job search whenever it's time to find your next step
• spend more time upskilling
• spend more time and money on your mental and physical health
• generally worry less day-to-day about food and other essentials for yourself and your dependents 

All of the above have massive benefits for your long-term impact and wellbeing.

Moving to a cheaper area might be particularly attractive for people who:

• want to explore high-risk/high-upside paths
• already have strong networks
• have kids or would like to have kids
• work in fields/roles with low-to-moderate salaries

You can (and should) still go to conferences and occasionally visit relevant hubs, especially when you're between roles, considering a career change, or raising money. It's fun, inspiring, and can lead to opportunities and connections. You want to stay on people's radar. 

But I think most people overestimate how much they benefit from living in an EA hub, especially after the first 1-3 months. This is hard to measure, but seems to true to me based on my own experience of having lived in Oxford for multiple years, having visited other hubs, and having observed other EAs who have lived in various places. (But please share your own experience in the comments!)

There definitely are exceptions, though. The additional cost of EA hubs in HCOL areas might pay off for people who:

• are early in their "EA careers", when they benefit loads from every (formal and informal) networking opportunity
• work in a team that is already in one location
• easily lose their motivation or experience value drift when they don't constantly surround themselves with EA community
• benefit hugely from slightly faster access to rumours on a particular localised community (e.g. some people in AI safety)

Lastly, don't move to the absolutely cheapest place your passport allows if that'll make you miserable. Ask around and try a few places. Consider coordinating with a few other people.

example

I personally live in Sheffield, England’s 4th largest city. I recently bought a 1-bedroom flat for £70k ($92k) in walking distance to the central train station. Sheffield has a walkable city centre, theatres, lots of small and large shops, access to nature, a cool arts/music scene, lots of vegan food, and good connections to other cities (e.g. ~2h train to London every ~30-60min).

I love London and probably like it ~50% more than Sheffield (mostly because of access to EA co-working spaces). But it’s ~3-6x more expensive on housing (depending on how much you're willing to compromise on convenience, and more still if you account for interest on buying a place, plus the extra rent you pay while saving for a deposit). I just can't justify moving there. I still visit for events, though.

My case is actually not a perfect example, as my exact current job probably couldn't be done from anywhere. But I intend to stay here even if this changes, at least for now.

miscellaneous related thoughts 

• Anecdotally, many of my EA friends in hubs don't actually "take advantage" of being in a hub that much. Many work from home most days and only go to two or fewer EA events per month. When they go to co-working spaces they often barely talk to anyone there.
• Also anecdotally (but mechanistically plausibly), groupthink seems more common in the biggest hubs.
• Building flourishing EA communities in cheaper (but still somewhat attractive) cities might
  • make the community more accessible
  • help with image issues that put people off EA, such as
    • elitism ("Only rich people from top unis are welcome.")
    • hypocrisy ("How can they claim to care about cost-effectiveness if they spend all their time in the most expensive parts of the country?")
• If your organisation pays people less if they do the same work from a cheaper area, consider if you are creating weird incentives. (See also this thread under this post.)
• Some related posts that some might find interesting or useful:
  • Moving to a hub, getting older, and heading home
  • Reflections on Anthropic and EA

I wrote this post myself but used Claude Opus 4.8 to critique the draft so I could iterate.

1. ^{^}

   I expect this to be the case for the majority of EAs, but if not please correct me and point me towards data that says otherwise!