There have been previous posts about the impact of founding a new GiveWell charity, the impact of charity founding, and some of the results of charities founded by EAs. This post, however, focuses on a specific question I get a lot. Namely, why should EAs in particular start charities? This is especially a concern for many EAs who are younger or inexperienced.


I think a strong piece of evidence for this is the historical track record of EAs founding high impact charities. Given that I already covered that in a previous post, I want to focus this one on why it's not just fluke or limited to extremely talented EAs, as well as the core reasons I think EAs can outperform other folks aiming to start nonprofits. My experience with this question comes from having worked in several charities founded by EAs and starting multiple ones myself.


Overview relative to other founders:


EAs will often be better at:

  • Cause and intervention selection
  • Value alignment
  • Networks and resources
  • Measurement and evaluation
  • Dedication and focus
  • Personal ability

EAs will often be worse at:

  • Experience working and managing broadly
  • Experience in the field
  • Connections within field
  • Related degree
  • Personal attachment to the cause

These are listed roughly in order of importance on each side. The rest of this post will explore each of these points in a bit more depth.


EAs will often be better at cause and intervention selection


The first point is one of the clearest. A huge amount of the impact rests on which specific charity gets created. Launching something aimed at being recommended by Animal Charity Evaluators or GiveWell leads to very different interventions than the vast majority of social endeavors. Most nonprofits focus on the personal experience of the founder, not on the most relevant data available. EAs are much more likely to use that data when making an initial charity selection. It’s not shocking that charities like Evidence Action or Charity Science Health go far in GiveWell’s system, as these sorts of charities have similar underlying goals of making the most possible impact per dollar using the available evidence. If EAs generally start fundamentally more high impact ideas, that goes a long way (although far from the whole way) towards having a large impact.


EAs will often be better at value alignment


It’s easy for any charity, even EA charities, to get mixed up about the goals the charity is trying to accomplish. Are we trying to get bigger as an organization or have more impact? Are we trying to increase the amount we help each individual or focus on preventing the most suffering as a whole? Questions like these come up very often in charities, particularly new ones. The different directions the charities end up taking can have a large impact on their endline effectiveness. For example, a chronic problem in many health charities is the overemphasis on helping fewer individuals more. Often other poverty organizations recommend Charity Science Health to do strategies that will increase our program’s impact on our already existing participants by 20% but at twice the cost. That same funding could have more substantial impact by reaching a larger number of beneficiaries as we scale. Although anyone can make the wrong choices on issues like these, EAs are far less likely to do so compared to typical charity entrepreneurs. In addition to this, EAs are more likely to hire other staff members who also think this way. For example, even my non-EA team members are more focused on cost-effectiveness than the average NGO employee.


EAs will often have better networks and resources


The EA community has access to a lot of resources that other charities struggle with. Things like early stage funding, social support, mentorship from others who have done similar projects, legal incubation, access to a value-aligned talent pool, and incubation programs are all very rare resources to have outside of the EA movement. Very few of these resources are field or intervention specific. For example, the help I was able to give to Fortify Health was not on the technical details of iron supplementation, but on the broad process of founding charities. This network and pool of resources is one of the things many entrepreneurs find difficult. Many nonprofits, including Charity Science and Fortify Health, would not have been able to get off the ground without this sort of support.


EAs will often have better measurement and evaluation


Getting team buy-in for a strong measurement and evaluation (M&E) system is a tricky process. Having EAs as the leaders of the organization greatly increases the chance of having quick feedback loops and the strong M&E required for a charity to be highly impactful. EAs will generally think of M&E-focused systems more naturally and be more emotionally invested in applying them across their charity. EAs are far from perfect on this, but it is a stronger part of EA culture than it is of standard animal advocacy, poverty or most other charity areas’ cultures. Many might think that M&E is best done by an expert who has experience at M&E focused organizations, and although I think these experts are very helpful to have on a broader team, I think there are too few of them willing and able to found charities to let them close all the gaps. Often an EA charity might work best with one of these employees hired (a hire many other charities might not make) to work on it full time, with the leadership also being M&E oriented.


EAs will often be more dedicated and focused


By “dedicated and focused” I mean being focused on the endline charitable goals and objectives. EA orgs that have this focus on doing good can be more flexible in making large-scale pivots towards impact. For example, New Incentives has had several large pivots, but because the founders are focused on and dedicated to preventing suffering (as well as being highly competent), they continue to be a promising charity. If an organization doesn’t have this dedication to the fundamental metrics, then it’s much more likely that a change of direction will mess up its focus on the end goal.


EAs will often have better personal ability


This point really comes down to what reference class you are comparing with. For example, if you compare to the baseline population of people founding charities, I expect EAs to do better than this reference class on metrics like getting into an Ivy League school. But I can see a case for drawing a much tighter reference class. EAs don’t generally want to start charities that are merely better than average, but charities that will be among the best. The reference class of “founders of the absolute best charities” (e.g. those recommended by Animal Charity Evaluators or GiveWell) is much more comparable to the group of EAs I expect to be able to found charities.


In the next section, I’ll explore more deeply the weaknesses that EA founders have.  


EAs will often be worse at having experience working and managing broadly


EA founders will generally be young and, importantly, have much less working and management experience than others who found charities. Although inexperienced entrepreneurs are common in tech startups, in charities it's still often people who have been working for many years. This management and general work experience is hard to teach via school or books and is often learned through personal experience. This is the biggest factor making EAs weaker founders, although I do think there are some slight mitigating factors. Small charities generally grow from a small team to a larger one instead of starting with a very large number of direct reports. EAs can hire more experienced middle management to help compensate, but even with this, there will still be a lot of rapid learning that will need to occur.


EAs will often be worse at having experience in the field


This is broken out from the other category above because one could, in theory, have work experience broadly, but not experience within the specific field. I think this lack of experience is many EAs’ number one concern. Unlike the above experience, this sort of experience is very niche. Fortunately that very fact is what makes it not that difficult to acquire because the area of specialization is so narrow you can learn it fairly quickly. For example, there are only a limited number of studies on SMS vaccine reminders in the developing world. As an Executive Director your knowledge set will need to be broader than just that, but you do not need to have a masters degree in public health to work on a global health intervention. You just need to gain a very strong understanding of the very narrow area you work in. My experience has been that EAs can generally get to this level of knowledge in a matter of months, not years, as many might expect, largely because the knowledge needed ends up being more specific than expected. The other common concern people have is that they won’t have the very local and cultural knowledge of the area. This is a fair point, but this is thinking from an individual instead of organizational perspective. You can, and should, hire local people who will be able to guide the organization on this front.


EAs will often have worse connections within a field


Some charities are started with heavy connections in the field, whether in the form of funders, advisors or co-founders. This network is often essential to having your charity run well. The reason I tend to rank this as less important is that I have found that it’s fairly straightforward to build up a network in your field of interest. For example, if you send out ten emails to field experts asking to speak to them about working in the area, between six to ten will respond, and some of those will be able to connect you to more people in the field (if you ask them to) and from that your network can grow quickly. Many nonprofits are quite happy to talk to others in the field and some of these contacts will end up being advisors and board members long term. I think that people generally expect that not many people will respond to a cold email, but unless the person is fairly famous, I have found that nonprofit people are very willing to do a quick Skype session. However, this approach does require more social skills than the slow build of working in a field for years.


EAs will often be worse at having a related degree


I think this factor is far less important for founding a charity than many others think it is. The domain-specific knowledge gained through a degree is less important than one might believe, for the same reasons given in the section on experience within a field. And as for  the reputational benefits of a degree, it's quite easy to hire a communication director or other staff members with degrees that provide that reputation, and in general this reputation will often quickly be overshadowed by the success of your charity. Think about top EA charities: how many of their founders have degrees in relevant fields? How many do you even know this information for? If you become very knowledgeable in a field, people are usually happy to talk to you and may often assume you have a degree based on your knowledge.


EAs will often be worse at personal attachment to the cause


This is the least important factor. Being personally attached to a cause area has the benefit of giving you more motivation for that cause area, but it also has the potential to cloud your judgement and lead to worse decision making.  


Overall, choosing good interventions and having good connections and resources overwhelm counter considerations, which are often less of a problem than anticipated, or can be overcome with intelligent effort or hiring. When considering these factors and the track record of EA-minded people starting strong charities, I think that this is a strong career path for EAs.

Sorted by Click to highlight new comments since:

On what grounds do you expect EAs to have better personal ability?

Something I've been idly concerned about in the past is the possibility that EAs might be systematically more ambitious than equivalently competent people, and thus at a given level of ambition, EAs would be systematically less competent. I don't have a huge amount of evidence for this being borne out in practice, but it's one not-so-implausible way that EA charity founders might be worse than average at the skills needed to found charities.

I thought your "dedicated and focused" section would be arguing that EAs would work harder, but it actually seems to carry on from your "value alignment" section. I'd suggest these should be combined into one section to avoid confusion.

A couple of comments:

  1. Could you state what your role and involvement is with various charities, and what those charities do, to provide some context? E.g. you mention helping fortify health but I'm not super familiar with what they do or how you helped them.

  2. Reading this, a worry I had is that new charities founded by would often by competing for the same pot of money from EA orgs and/or individual EAs. Do you think this is likely to be a problem? It seems the success of this strategy relies on Open Phil do a lot of the funding. If new EA charities instead raise money from ineffective charities (possible), or raise money from people who would not have donated (not that likely) then this isn't a problem.

1) I hope to publish a post soon specifically going into the help I gave fortify health and what help I can give future charities, but I can clarify briefly here. Charity Science Health - I was on the research team to pick the intervention + cofounded + worked full time in a co-ED position for the first 2 years of its existence. Effectively I was involved as much as one could be in a charity. Fortify Health - I was on the research team to pick the intervention + connected the co-founders when one reached out to me + Gave them a seed grant for their first 6 months + helped them in a consulting role ~5 hours a week over those 6 months. Effectively I was like a highly involved board member.

2) I think this is a huge concern, I generally think EA charities should be aiming to be the highest impact charity in a given field. E.g. a lot of the value of CSH comes from the small chance we can be higher impact than AMF. If CSH for example fell between the effectiveness of GD and AMF, CSH would pretty aggressively try to seek funding outside of the EA community (including GW/OPP). This partly to do with “the last dollar spent” in poverty likely being pretty high impact (see this post on talent gaps for more details). In something like AR, given the funding situation I think the more important consideration would be whether a new charity has a good chance of beating the bottom 25% of charities funded by OPP/ACE.

Connections in the field seems to be quite an important foundational issue, but whilst it may be a weak area generally, i think it can be an area where insufficient time is spent considering the importance of plurality. So if a certain group of people were asked to be part of the experts in the field then it could become fairly self recommending from there on in, particularly if it were resourced / various benefits flowed from it. I tend to view this as a bit of an issue within EAA, particularly at both ACE and the Open Philanthropy Project where approaches have a tendency to not be given equal consideration, instead some are valued highly (particularly those aligned with direct utilitarianism) over others.

I think this can then lead to other issues in terms of internal evaluation. So in-group bias wouldn't be challenged because external evaluation has been devalued. Creating a bit of a problematic loop.

More from Joey
Curated and popular this week
Relevant opportunities