Creating a low cost of living EA hub + co-living/co-working space for people trying to effectively change the world for the better seems, to me and many people I've spoken to, like a potentially unusually high value project. This is something I'm personally very interested in (though I've been prioritizing other projects recently), with the plan which seems most promising to me being buying and converting a hotel (with people who want more independant living space renting nearby). Simply picking a location as a schelling point is also an option, but having a place for new arrivals to easily get involved and a founding group moving together to focus on setting up a specific location seems likely to increase the chance of hitting critical mass.

I've created a spreadsheet of countries which have been suggested at some point or seem like they have potential from a glance, and would like to open up information gathering to other interested EAs, so we can to help narrow down the large list and start looking at a few specific locations. Helping out with this is something which anyone can do in a spare few minutes (or suggest to people who want a flexibly sized EA task), pick a box to fill in, check Nomadlist, Teleport, or google relevant things, read, add summary+useful links.

Additionally, Geordie and I created a survey to gauge interest and learn what people want from a new hub. I'd love some feedback on the survey itself (especially from people who've done surveys before and know the kinds of things that we'll probably wish we asked) before we try and widely publiscize it.

Suggestions for names and other countries which are worth investigating are also welcome.

Assorted links for background:

Facebook group

Slack (send me your email for an invite)

Hackpad

Discussion on FB

Older discussion on FB

Trello

29 comments, sorted by Highlighting new comments since Today at 2:56 AM
New Comment

I'm glad you're doing work on this - it's a potentially very valuable project. I think we could go about it in a different way though. There's a risk of analysis paralysis in trying to find the optimal location in advance so that we can commit to something as big as buying and converting property. Instead we could just find the people who are likely to move somewhere cheaper in the next few months (I'm one of those people) and see if we can do it together. We might also want to drop the framing of it as 'A new EA hub' at this stage because that makes the task seem big, important, and intimidating. Let's just experiment with some locations and see how it goes. We'll learn something about living abroad and we'll be able to observe existing coworking and coliving setups to see what works.

Agree.

Also, the 80,000 Hours team lived in Chiang Mai for 6 weeks, and that seems like a great option. I could also see us going back 1 month per year or so.

I think the approach you're talking about here is sensible and would be a good way to start things if you're wanting to move relatively soon.

The reasons I'm mostly thinking about this approach instead is partly that I don't expect to be in a position to move in the near future (minimum six months, probably somewhat more), meaning planning does not delay my direct involvement, partly because I suspect it's more likely to hit critical mass and start attracting more people rather than staying small/dissipating if an initial group go over together to set things up and provide a seed community/safe&known arrival point, and partly that picking a better location seems high-value and it's easier to gather the initial info online than send people to different countries.

That makes sense, you're not preventing your own moving by doing the analysis as you have other reasons for not moving yet.

Can I suggest an amalgamation of our approaches then:

Phase 1: Exploration. In this phase, those that can move in the next 4 months move to a location that would be good for them and try to join together with other EAs in doing this. They also try to explore more than one location and report back their findings to the whole group. Those that can't move that soon but are interested in the idea can contribute through online research. Everyone can help those who are interested in moving with location choice.

Phase 2: Clumping. In this phase, we take the findings from phase 1 and choose one (or a few) standout locations to concentrate on. We encourage more people to move there, including EAs that have gone to other locations.

Phase 3: Community-building. Once we've got a group of > 15 people we can start to invest in community-building projects such as coliving and coworking spaces and outreach to the local community.

Each of these phases is useful even if it doesn't progress to the next phase.

This approach gets the early adopters moving and gathering useful information whilst also creating the seed group effect that could attract more people in the future.

hm, depending on how many locations are explored in the next few months we may or may not be confident we're in one of the top few locations (imagine a world where, three or four places are looked at, one of them stands out but based on other research we suspect that there are other countries which would be significantly better to establish a hub for administrative reasons. In that case I'd be in favor of trying to visit the other countries before moving to stage 2.), but in general that plan makes sense. I like the idea of collecting info on the ground if you're keen to get started soon, do you know of others who are ready in that timeframe.

I also think that >15 would be the number of people who would be there on average in expectation (so, a significantly higher number expressing strong interest/commitment), rather than people willing to do initial setup work before a co-living location is ready. The founding group could plausibly be somewhat smaller (~4-10?).

If this ends up not hitting critical mass naturally (but there is still strong interest), then we can likely give it a boost by moving a founding group willing to work on setup of the co-living space in one go.

I think the "analysis paralysis" objection is exaggerated. Even if you run it as an experiment, you need to put a lot of thought into where you decide to run that experiment. It's unlikely that you'll test very many places, so you'd better do some thinking in advance.

I agree that you have to do some thinking in advance - you have to choose at least one place to go. However, I don't think this is a very hard a choice for someone to make because the digital nomad scene has already identified a handful of good places. From my reading of recommended places in digital nomad forums, here are the places that stand out for cutting your living costs whilst doing remote internet-based work if you are from a Western country:

  • Chiang Mai, Thailand
  • Ubud, Bali, Indonesia
  • Medellín, Colombia
  • Prague, Czech Republic
  • Budapest, Hungary
  • Las Palmas, Canary Islands, Spain

There are only a few of them, and personal preferences will play a big role in which of them each person would prefer. Each person who is seriously interested in being part of this project can choose the location that's best for them from this or similar lists, and then report back about how it's going once they're there.

My plan is to go to one of the European locations this summer. And if it doesn't work out, I can always go somewhere else.

I'd be interested in hearing arguments for why a hub in a low-cost country would be better than a new hub in a developed country with more potential EAs (e.g. Australia, the East Coast).

Primarily much lower property and living costs, meaning people can live there for less (= more donations, longer runways for startups, lower expenses for researchers, etc) while still retaining high quality of living and being around interesting people. Hubs in higher cost countries would likely be valuable as well, but they cater to a different group of people, would require higher initial investment for a comparable property, and generally have stricter visa requirements.

It's funny that people forget how cheap most of the U.S. is. In a town near where I live, a Victorian mansion with >40 rooms was for sale for $500K. And it comes with all the benefits of living in a developed country...

nods, a hub in a low-cost part of a developed country is also worth considering, but I think it'd still be significantly higher cost than a low cost country (food, assistance, and building), and has the disadvantage of being significantly less visa friendly (the US in particular puts major hurdles in the way of non-US citizens). The really low cost locations in the US are also likely to be less attractive locations, in terms of access to major cities and environment.

I also think that being in a developed country may be less of an advantage than some people are imagining, because I expect the internal and online interactions to be much more significant than local interactions once we're outside existing hubs and other high-cost areas.

I think despite these considerations it's worth exploring and also a potentially high value project, but it's not something I'm drawn towards.

I don't think this counts as sufficient reason. You need to list all of the (major) pro-and-con-considerations and weigh them up against each other in order to show that a low-cost EA hub is better than another one in a developed country.

Evan's comments below are interesting, though.

It's not sufficient reason to think that low-cost hubs are higher value than additional high-cost hubs, but it is (I think) sufficient reason to think that whether or not EAs are working towards more hubs in higher cost countries, it would still be good to also be moving towards having at least one low-cost hub.

I have greater personal interest in low cost hubs, and I suspect that getting the first low-cost hub will be unusually high value (for reasons in some of the discussions on facebook/in person which I will write up properly at some point), so that's what I'm focusing on here. I do not intend to crowd out or compete with efforts towards more high-cost locations, I see those as having a quite different set of benefits which I'm not very confident in my ability to evaluate the impact of (since it depends massively on locations I don't know at all well, unlike the low-cost one which would attract people from across the globe). I would be happy to disambiguate by using a different name if the people working towards high-cost hubs want "New EA Hub".

I suspect these don't trade off against each other meaningfully.

Might make sense to get a bunch of people to precommit to coming if at least N people come, like the Free State Project.

In the first instance, you mightn't necessarily need a lot of people to come, but if you wanted to grow it to 20 or more, that would be an interesting way to grow interest.

Yes, this is definitely something we'd want to do. I've added a question to the survey about how large a community people would be happy to move into, and will either work out how to usefully include precommitment in the survey at this stage or do that as a separate commitment once we have contact details for everyone interested and know more about where and how we'd do this.

Cool. Unless you're planning to move >20 people all at once (insanity), I figure it's better to pitch and discuss through personal correspondence rather than a pro-forma precommitment.

Once people have actually moved, you'll get more widespread interest, so you can create a form where people register to move if there are >N people, where N is 50 or 500.

By that point, you're basically overtly founding a new radical activist community, though, so you'd better get good at careful public relations.

nods, makes sense. My current guess is that at least initially most people would spend 3 months to a year visiting while working on something specific rather than moving as permanent residents.

I'd be aiming less for an activist community in the country, more a tech/startup/research hub for people with an interest in fixing global problems. I don't foresee a focus on influencing the host country. Good point that perceptions around that are important though. I'm very open to advise and assistance from people with PR skills on how to present this.

https://teleport.org is another source of data on which cities to move to, similar to nomadlist.

Apparently this paper found that 36% of successful changes in behavior were associated with a move to a new place (nearly three times the rate associated with unsuccessful changes)? So it might be worthwhile for participants in this project to plan out some habits they wish to form in the new place in advance of the move? Relevant thread.

Yes, this class of self-improvement seems like one of the larger potential wins, especially if there's a supportive community in place to help.

I've been thinking of writing an article on the EA Forum for a couple months now, something of a companion article to this one, about cities which have the potential to become EA hubs. This is a different angle of attack than the thinking which has been poured into EA hubs which have are explicitly researched for their low cost-of-living potential. There are a few cities around the world which have the potential to become EA hubs in the same way Oxford or SF are, even if they're not as advantageous as ones in countries with drastically lower living costs. Still, though, it can't be anything but useful to be at least be aware of these options. Other things being equal, as an ecosystem, if EA could or would benefit from a greater diversity of hubs to set roots in, it's good to be aware of that.

AFAICT, the extant, bona fide EA hubs are:

  • Oxford (with offshoots in London and Cambridge)
  • Basel, Switzerland
  • the Bay Area (albeit a large area where their are nuclei around Stanford, Berkeley/Oakland, and San Francisco itself, each of which are at least 1-hour transport away from each other)
  • Melbourne

Other potential EA hubs include:

  • Boston, Massachusettes
  • Vancouver, Canada (where I live)
  • Berlin
  • Seattle, Washington

I imagine there are a few more which would qualify. Let me explain what I mean by 'qualify'. Ideally, what I would do is determine what are the common factors between SF, Oxford, Basel, and Melbourne. These would be factors that would, if not directly precipitate the generation of an EA hub, be highly correlated with its rise. Then, I would figure out which other cities share those factors, and speculate/encourage those cities to develop other common factors, such that their potential to become an EA hub is more robust. Such factors include:

  • at least one, and usually multiple, EA organizations
  • the organization of at least one, often multiple, local meetups which themselves sport ~10 regular attendees, outside of the local EA orgs
  • sharehousing between EAs, or otherwise factors which facilitate incidental social cohesion among otherwise professional relationships
  • a local culture amenable to either startups and/or social entrepreneurship, such that the city itself as an attractor for the sorts of people drawn as EA, and the city can feed the growth of the EA hub.

The biggest problem here is that whatever it is in the culture of hip cities which draw socially conscientious millenials, that lends themselves to becoming EA hubs, also draws enough people that it makes them very expensive. London is the most expensive city in England, and I'm guessing for a university town, Oxford isn't too cheap either. SF is the most expensive city in the U.S. I hear Melbourne is one of the most expensive cities in Australia. Vancouver is the most expensive city in Canada. Maybe only second to New York, Boston is one of the costliest cities to live in on the American east coast. I know little of Basel, but based on the above, I wouldn't be at all surprised if it's expensive to live in.

I think i know what's going on. EA hubs tend to be cities full of people open-minded to experimenting with cutting-edge innovation. Effective altruism is just the mindset of 'cutting-edge thinking' applied to philanthropy and do-gooding. To get a city with cutting-edge innovation and risk-neutral young people, though, comes with high costs of living. This isn't exactly a scientific theory of how EA hubs rise and fall, but I think the relationships here are pretty clear-cut.

The curse of founding a new EA hub is that the places, the local cultures, in which one would ideally want to seed an EA hub are also the most expensive ones. If an exodus of EAs moved to a new city in the developing world, bereft of the 'startup culture' mindset, we might have to build that culture from the ground up, all by ourselves. That seems like it would have high fixed costs, and be very hard to do.

Here are a few data sources for finding cities with a culture or sub-culture that has EA-potential:

This is very similar to my model of the highly active EA places are expensive problem.

I suspect that creating the culture from the ground up would not be extremely hard, though it would impose a fixed cost/require some dedicated people to push it forward. Primarily because we could bring people in who are already familiar with relevant mindsets/ideas rather than having to spread them to locals where we arrive. I would expect a group on the order of ~15-30 awesome people (+low-cost, +high quality of life) to start to be attractive to many from around the world, even if the wider culture of the country/city was not particularly involved in 'cutting-edge thinking'. To start with only those okay with a relatively small/tight social circle would join, but as it grew people who wanted a wider circle of interaction would be happier. Most people don't need a huge social circle, so long as the wider culture is not actively hostile towards us and we were careful to build a healthy internal social environment early on I think it'd be okay.

This is quite a different project to setting up in a place which already has a bunch of EAs, but it seems doable to me. Rough steps: figure out what people want, gather names, and pick a location, then find funding, scout the location, and pick an exact building, then bring a founding team over (~4-10 people?) to do initial setup, then start bringing more people who want to be part of that kind of group over to help and live. Once you have a seed community and a schelling point for interacting with it expansion is relatively straightforward (people rent nearby, similar co-living spaces pop up, etc).

Totally in favour of exploring this sapce. I've been paying attention to the experience of EAs who are from or who have lived in South American countries over the last couple years, and there seem some promising points of inquiry. I've written down what I think about opportunities in Argentina, Brazil, and Chile below.

Argentina

Just pointing out that after I heard about how easy it is to acquire visas to work in Argentina, I'm far from the only one that thought this would be a natural candidate for a low-cost EA hub. I notice in the spreadsheet you mentioned a major disadvantage might be economic/political instability in Argentina. I know things tanked there in the late '90s, and I don't know how things are know. Definitely worth researching more to develop a model of how problematic it may or may not be over the course of several years. I'm guessing with Argentina EA could send a contingent of, say, a dozen EAs willing to try living there as expats for a couple years, letting us know how it goes and getting better 'inside view' data on how likely catastrophic disruption to their lifestyles seems, and then reporting back to us. I'm aware the whole point of an EA hub is to prior determine a great location for dozens, perhaps hundreds, of EAs to lay roots there for at least several years on end. However, the potential advantages of Argentina, presuming there isn't a political/economic crisis there, seem to me to great to write off. So, I think it'd be worth trialing with an experiment amongst ourselves. Alternatively, EA could develop an action plan such that if things get hairy in Argentina we can move back to other countries quickly, such that we haven't laid down roots or sunk costs in Argentina that make it hard to leave.

Brazil

Diego Caleiro, Joao Fabiano, and Leo Arruda are three guys I know who are or were highly involved in IEFHR (Instituto por Eticas, Futuro de Humanidae y Racionalidae; English: Institute for Ethics, Future of Humanity and Rationality) in Brazil. The fact there is an EA org in Brazil might be a good way to get a foot in the door. Diego Caleiro is himself an open-minded guy who is in favour of independent researchers in effective altruism receiving funding So, I think he'd be open to EAs being hired by IEFHR as employees, and then getting paid to do whatever work they would normally do. This might be a way for several dozen EAs to work out of Brazil. Of course, the money would need to be doanted to IEFHR for this to happen, which might be a minor technicality. I don't know if donations to IEFHR from outside Brazil would be tax-deductible in the country of origin, so one could do a CBA of the transaction costs of donating to a Brazilian org instead of, say, an American or British EA org, plus the costs of moving and getting settled in Brazil for a given individual, minus the reduction in living costs. If the sign came out highly positive, it might be worth people moving to Brazil to work at IEFHR.

I don't actually know how open IEFHR would be to this sort of arrangement. At most, I would think they would want concessions from other EA orgs or the EA community-at-large to plug their preferred causes, e.g., far-future ethics and FHI-style cause prioritization. Generally, I imagine they're pretty open to free cooperation with the rest of EA.

The bottleneck here might just be the bureaucracy of Brazil. I know the folks who run the Charity Science Foundation of Canada (CSFC), the legal parent organization of EA orgs Charity Science (CS), Charity Entrepreneurship (separate project/org run by the founding team of CS)and .impact. It's legally incorporated as a foundation, as opposed to a run-of-the-mill registered charity. The upshot is that having the legal status of 'charitable *foundation' as opposed to just any 'charity' is that certain advantages are granted. Some of these might be ease of hiring foreigners for your organization. For example, I'm aware that, combined, the three organizations headed under CSFC have helped acquire Canadian visas for, and hired, 6 foreigners (3 Brits, 3 Americans).

I know the status of 'charitable foundation' exists in the United States and the U.K. as well. I don't know if Brazil has a similar set-up. If IEFHR doesn't have an equivalent/comparable status of 'charitable foundation' in Brazil, it might be exceedingly difficult for them to get foreigner-EAs approved as employees, no matter how much money is moved through IEFHR. I honestly don't know enough about civil services in Brazil to comment. I'll ask Leo about this.

Anyway, if you, Eric, yourself, want to get more info on the potential of using IEFHR as a lever to make an EA hub in Brazil, I recommend talking to Joao. IIRC, he's currently a fellow at FHI working with Anders Samberg. You should be able to meet in-person, or Skype, with him pretty easily.

Chile

Brienne Yudkowsky spent a winter in Chile. The rationale was more or less that anywhere she wanted to live in the U.S. was sufficiently temperate and cloudy there was too-high a chance she'd be hit by seasonal affective disorder (SAD), and she wouldn't tolerate that another year, so she spent a few months in the tropics. I recall Eliezer once mused on FB, other things being equal, and the Bay Area not having comparative advantages for EA work, for everyone to migrate to a country like Chile to form a new EA/rationalist hub. I generally gather Brienne had a positive experience in Chile. I don't know more, but you can ask her yourself.

Hello guys. I believe I can shine a light about the state of things in Brazil.

First, about Brazil in general, it is a big tropical country (200m) with sunlight the whole year, very hot at north, cold at south (but not to the point of snowing). We are in a politically unstable moment due to enormous corruption scandals and political dissatisfaction involving both the government and the opposing parties (check the news, it is a mess), but in IMHO this won't lead to any sort of dictatorship nor utterly compromise the economy although we are facing a depression that might last for a decade. Exchange rates are very favorable to foreign money (1 US dollar = 3.5 Brazilian real). We speak Portuguese, though I believe most upper-class and upper middle class are able to speak English. Brazil is a developing country with poverty issues (specially at the north and north-east), bad public education and tens of thousands NGOs (most of them probably not very effective though). I can't give you a picture of the IT sector but I think it is probably growing well. We have reasonably advanced academic research in many areas including biotech but the political climate is unfavorable at the moment (some scientists are even moving to other countries). I live in São Paulo, a huge city (12m) with mild climate (10-30°C )and Brazil's main commercial center, we have a small but growing EA community (I'd say about 8 core members plus 15 others) and meet up monthly. We are currently studying charity evaluation to check how effective are social projects in Brazil. There are also EAs in Belo Horizonte (talk to Celso Vieira about it) but I don't think there are regular meetups there or anywhere else.

From my limited observation I think US work culture is much more result, efficiency, accountability oriented than ours. I believe this translates also in less support for EA, since people are much more prone to an emotional approach to charity. Also we are more to the left in the political spectrum on average, and people are much more prone to appeal to state intervention and social programs to alleviate poverty, they are also more open to marxist ideas and have a weaker culture of philanthropy.

About IERFH.org (Instituto Ética, Racionalidade e Futuro da Humanidade, Institute for Ethics, Rationality and Future of Humanity - Evan please correct it), I'm currently a director and we are not currently officially registered (mainly for bureaucratic reasons, although we could easily do it if it was important). We are mainly focusing on promoting rationality and transhumanism for the moment since the Brazilian EA movement is growing by itself. But of course we would be willing to help and to promote support for any EA projects.

Have added links to this to the relevant column in the spreadsheet. Thanks for the leads!

This is interesting. What do you think about including in the consideration set the ability for this community's location to have a positive impact on the world? For example, moving to a poor area in need of external income or an area where westerns have a negative bias? There could be some secondary and tertiary benefits to this community.

I'd imagine there would be effects of this kind, but I expect the local impact to be much smaller than the impact flowing through enabling people trying to improve the world to be more effective and live on a tighter budget, so I'd not want to prioritize optimizing this class of effect over picking a place that works best for EAs. Still, if there are two otherwise similar locations, one of which would likely lead to notably larger effects of this kind, it could be a tiebreaker.