The GiveWell Blog

Inaugural open thread

A commenter recently suggested we try out a periodic open thread. Let’s give it a shot!

Our goal is to give blog readers an opportunity to publicly raise comments or questions about GiveWell or related topics (in the comments section below). As always, you’re also welcome to email us at if there’s feedback or questions you’d prefer to discuss privately. We’ll try to respond promptly to questions or comments.

If this is helpful, we’ll continue to do them periodically in the future.


  • Alex Cuevas on November 14, 2014 at 5:55 pm said:

    Have you ever considered doing a podcast?

    If you have considered it, why did you reject it?

    If you haven’t considered it, is it something you’d be interested in pursuing down the road?

    Other GiveWell blog readers, if there was a GiveWell podcast, would you follow it?

    Podcasting is becoming a bigger and bigger source of media / information for people, myself included, and I think it’s the ideal media format for GiveWell-like information and discussion (other than the blog form that it already inhabits). If there were a GiveWell podcast I would probably listen to every episode and I bet a lot of other EA’s and non EA’s alike would too.

  • Peter Hurford on November 14, 2014 at 10:50 pm said:

    In the spirit of an Open Thread, anyone got any guesses as to the upcoming changes in recommendations? 🙂

  • Kevin Burke on November 15, 2014 at 12:50 am said:

    when you do conversations, are there certain questions you like asking to everyone? what questions/responses tend to distinguish good organizations from bad?

  • grendelkhan on November 15, 2014 at 4:03 am said:

    Are there any plans to evaluate promising avenues of underserved research? For example, broad-spectrum antivirals (PLoS One article here); some people have set themselves up to take donations, which makes this particularly interesting for me.

    I suppose it’s like picking one’s own stocks for investing, but I’d be curious if the people here have any thoughts on the matter.

  • Zeren Ozdamar on November 15, 2014 at 10:12 am said:

    Among Givewell`s criteria, one is a focus on evidence-backed interventions.

    Is there any detailed explanation (or a criteria) of what is considered a valid goal by Givewell researchers?

    For example, the goal of saving a life, in fight against malaria is naturally considered a valid goal but the rest would be related to quality of life and this concept is less clearly defined.

    Increasing income levels, for instance, seems to be considered a valid goal as well, perhaps based on the fact that it is a factor in Standard of Living.

    Which other measures can be used to define valid goals?
    (Human Development Index, Physical Quality of Life Index etc.)

  • Taymon A. Beal on November 15, 2014 at 11:37 am said:

    Zeren Ozdamar: You might want to have a look at this post:

  • Does GiveWell ever directly seek out big donors to encourage to give to their recommended charities? Why or why not?

    It seems that there could be some argument for seeking out a small number of big donors rather than making general recommendations for a large number of small donors in terms of overall effectiveness and money moved.

  • I have many questions. Great idea.

    1. Where does GiveWell fall on the aid v. foreign investment debate? Has GiveWell established an opinion that well-given aid is helpful, and better than investment?

    2. What sort of positions are available at GiveWell that aren’t open at the moment (mid-level to senior)? Will any be open soon? What sort of background/experience excites you? Are post-graduate degrees 100% necessary?

    3. Are there any opportunities to get involved with the Open Philanthropy Project?

  • Tony Scott on November 16, 2014 at 2:49 am said:

    Nice idea. Why is the number of top charities 3 (opposed to 2 or 4, for instance)?

  • Maybe a bit of an aside, but what are the ethical considerations for Randomized Controlled Trials such as the RCT done with GiveDirectly? Is there any body that oversees such ethical considerations and approves or rejects trials, such as ethics boards for psychology experiments?

  • Alex Cuevas, no, I wouldn’t listen to a podcast. They’re horribly inefficient. Most hour-long podcasts can be compressed into a two-page essay with no real loss of data or detail.

    Duff, from my understanding, GiveWell does do direct fundraising from big donors. However, it seems the fundraising is for the organization itself. My guess is GiveWell thinks GiveWell itself is one of the best giving opportunities today, though they won’t go out and advertise themselves as such to the broader public. (Someone from GiveWell correct me if I’m wrong.)

    RE: Ethical considerations of RCT for GiveDirectly. I can’t imagine there being many big issues. Giving a big wad of money is about the least harmful thing you can do to people that poor. What harms there are are clearly outweighed by the benefits of learning.

    Here’s a question for GiveWell: What happens to former top picks? I’m interested in, say, the experience of VillageReach and AMF. Does the money dry up or is there a lasting stream of sticky revenue?

  • Zeren Ozdamar on November 16, 2014 at 1:51 pm said:

    Taymon A. Beal, thank you for pointing me to the blog post on value judgements.

    I found answers to my question in the following two sentences from that post:

    -The primary things we value are reducing suffering and tragic death and improving humans’ control over their lives and self-actualization.

    -We believe that further economic development, and general human empowerment, is likely to be substantially net positive, and that it is likely to lead to improvement on many dimensions in unexpected ways.

  • I would be interested in hearing any updates/timelines on when OPP and GW will become more separate organizations (if this is still the plan).

  • Alexander on November 18, 2014 at 2:35 am said:

    Hi all, thanks for the questions and comments!

    In order:

    • Alex: We haven’t considered starting a podcast in much depth before. We’d guess that the work involved in planning, recording and editing episodes would be pretty significant and we don’t think it’s very likely to be a better use of staff time than our current (relatively limited) outreach activities, but we’ll think about it something we could do as part of an increased focus on outreach/marketing.
    • Kevin: We don’t have a straightforward template of questions that we ask organizations we’re considering for a recommendation. We tend to focus on trying to understand the process from donations to impact in as much detail as we can, which is often revealing. We try to understand as much as we can about how the organization’s model could fail to translate additional funds into impact, and to answer other questions about our criteria, like room for funding or transparency.
    • grendelkhan: We’re planning to do more work on funding for scientific research next year, which could include relatively applied research on things like broad-spectrum antivirals, but also potentially much more basic research (such as new tools and techniques).
    • Duff: Yes, we do some targeted outreach to large donors. We tend to think in terms of trying to maximize the “money moved” rather than the number of donors, and the distribution is very skewed: most donations in dollar terms come from a very small portion of all donors. We write about this in our annual review.
    • Josh:
      • We haven’t rigorously assessed the question of social returns to for-profit financial investment in low-income countries. I’d generally expect them to be small because I’d expect investors to capture the bulk of financial returns. However, we’ve focused on assessing charitable donations rather than financial investments.
      • We aren’t currently recruiting for more senior roles, but it’s possible that we could be in the future. Postgraduate degrees are definitely not a requirement. There’s no particular set of backgrounds or experiences that we’re currently looking for, but if you’d like to talk, feel free to email us.
      • I’m not sure what you mean by “involved.” Right now, there aren’t any particular opportunities to get involved with the Open Philanthropy Project other than working at GiveWell.
    • Tony: We didn’t make a conscious choice to have 3 top charities: we felt that there was a clear cutoff in quality between our current top 3 charities and the other ones we might consider. We hope to have more than 3 top charities in the future.
    • Duff: I’m not sure, but I’d guess that the academic researchers who conduct the RCTs in collaboration with GiveDirectly do submit their projects to Institutional Review Boards, the same institutions that are responsible for reviewing psychology experiments. But I also think that GiveDirectly’s randomization represents minimal ethical concerns, so I’m not particularly concerned either way.
    • Sam:
      • I wouldn’t agree with your characterization about our fundraising – we do try to solicit donations to our top charities, not just for GiveWell’s operating costs.
      • Our impression is that donors giving based on our research tend to be loyal to our recommendations rather than the charities we recommend. For VillageReach, we would guess that relatively little funding has been sticky (less than a few hundred thousand dollars); AMF is more complicated because of the room for more funding considerations that led us to de-list them last year. We last published an update about VillageReach in February 2013 and are planning to continue to issue updates on AMF.
    • Joey: We still don’t have a clear timeline; I wouldn’t anticipate formal or legal separation until after 2015.
  • Uri Katz on November 18, 2014 at 6:50 am said:


    Do you have information on how a donor can increase the money they donate, through smart accounting practices and the like. I recently learned 2 very interesting facts, which leads me to believe there must be others: 1) when one donates stocks and the NPO sells them, the NPO does not pay tax on the profit; 2) many large companies have some gift matching plans for their employees (I read your article on the issue, and while you discourage most donation matching, you encourage it when the donor can choose from any tax exempt organization, and the money will not go to charity otherwise).

    Considering that these two facts alone would have increased my giving to date by at least 110% (I am being conservative since I know there are overheads and commissions involved), with little effort on my part, I think such information should be linked in a prominent location on your site.

  • Matthew Dahlhausen on November 18, 2014 at 11:33 am said:

    Hi –
    Would it be possible to create a stack-exchange or reddit style forum for GiveWell? This open thread seems highly disjointed.

    Effective Altruism at University of Maryland identified an area of inquiry:
    – LivingGoods is in the running for being a top-recommend charity this year
    – Cookstoves are a significant amount of the ($) sales, (22% in Uganda, 56% in Kenya)
    – The benefits of cookstoves – reduced fuel use, cooking time, and air pollution – vary considerably by the type of stove.
    – LivingGoods sells improved biomass stoves.
    – There is good evidence to suggest that these stoves do not reduce emissions enough to reduce disease burden from particulate pollution.
    – They may still be very helpful (people buy them).

    We recommend GiveWell work with LivingGoods to see if it is appropriate to change the cookstove offered, or to see if offering different types of cookstoves would improve outcomes. Dr. Kirk Smith at U.C. Berkeley is an expert on cookstoves, and it would be worthwhile for GiveWell to engage him to do a more in-depth review of the cause.

    We’ve created a sort presentation on air quality in poor countries, looking at cookstove impact, and air quality concerns in the U.S., focusing on residential air quality (it also has a lot to do with stoves). It contains sources for the statements made above:
    (link may not last. sorry. such is the digital age.)

  • Ryan Carey on November 18, 2014 at 7:35 pm said:

    Hi Matthew,

    Although there’s no Eeddit or Stack Overflow for GiveWell, there is an Effective Altruism Forum, which uses the Reddit codebase.

    I’ve been a part of building the forum, which incidentally links to all recent GiveWell posts in its right-sidebar.

    Its organisation helps with in-depth conversations and so it should complement ordinary blogs on GiveWell and elsewhere.

  • Alexander on November 18, 2014 at 7:59 pm said:

    Uri – Thanks for the suggestion. We think this information is of considerable value for some donors (i.e., those with company matching programs and/or appreciated stock), but that such donors are a relatively small minority of our audience. Nonetheless, we’ll think try to mention these practices more frequently or prominently (our last blog post on the topic was two years ago). And for what it’s worth, I don’t think there are other major opportunities that you’re missing!

    Matthew – Thanks for the comments and suggestions.

    • I agree that our comment structure isn’t ideal for lots of back and forth like this thread; if we continue to have a lot of threads like this, we’ll consider changing it. Alternatively, we could also host open threads on the EA Forum that Ryan mentioned.
    • I appreciate your presentation on cookstoves. Most of our assessment of Living Goods to date has been more focused on the goods they distribute that we believe have been responsible for their impact on child mortality, but we agree that there’s room for us to have a better understanding of the health and economic impacts of LG’s cookstoves, and that’s one of our major questions for further investigation.
  • Nicholas on November 18, 2014 at 11:43 pm said:

    Do you have any charity recommendations for donors who primarily want to make painkillers, anaesthetics, less painful surgeries, and so on, available to the poor?

  • Alexander on November 19, 2014 at 10:29 pm said:

    Nicholas – no, sorry, we don’t have any recommendations for that focus.

  • Ian Turner on November 23, 2014 at 9:30 pm said:

    Sam, I disagree about the possibility of harm from cash transfers. For example, one might suppose that cash transfers could lead to conflict, violence, laziness, drug use, or other bad effects. I think at this point we have evidence that such outcomes are minimal, but I’m not sure we could have known that before having conducted any trials.

    For an extreme (fictional) example of how one can be ruined by access to resources, read John Steinbeck’s short novel “The Pearl”.

  • Hello,

    Do you (or anyone else) happen to know whether any particular charity is preferable when it comes to the Syria Winter Crisis? As reported here

    – Thanks!

  • Alexander on December 1, 2014 at 10:06 pm said:

    Holly – no, sorry, we don’t have any recommendations focusing on the crisis in Syria. We’ve shared some general thoughts on disaster relief giving here.

  • I searched your site for “givingtuesday” but didn’t find anything. A blog post about your opinion of this relatively new idea ( would be interesting. I am seeing this year that some organizations (PIH, for example) are offering 1:1 matching for today only.

  • Alexander Gabriel on December 7, 2014 at 11:07 pm said:

    I see a problem with arguments for acting against “AI risk”: our total ignorance of crowdedness between today and whenever AI was discovered. Technological advance is an evolutionary process, so clearly any future leading up to human-level AI would be very volatile. It seems hard to say anything about global politics then or which causes people would be working on.

    Does this strike you as a major issue? I see crowdedness as on equal footing with importance for the attractiveness of causes in general, and don’t see why crowdedness would matter less with AI. Does this seem reasonable?

  • Alexander on January 7, 2015 at 2:26 pm said:

    Justin – sorry for the delayed reply, and thanks for the suggestion. We’ve experimented with some kinds of Giving Tuesday engagement, mostly notably in 2013 when Good Ventures announced that they would match contributions to GiveDirectly 1:1 between Giving Tuesday and the end of the year (up to a $5 million cap). At this time, we don’t see more engagement around Giving Tuesday as being the best use of our current outreach efforts, but that could change, especially if we decide to invest more work overall in that area.

  • Holden on January 8, 2015 at 6:53 pm said:

    Apologies for the delay in responding, Alexander. I agree that crowdedness is important and that the future crowdedness of working on AI is unknown, but if crowdedness today is low, and if opportunities exist today to do important work that can’t be substituted for by future work (e.g., perhaps obtaining the necessary insights is a multiple-decade project, and starting today rather than 10 years from now would make a crucial difference), it could still be an attractive cause today.

  • Richard Chappell on January 30, 2015 at 12:16 pm said:

    There’s an interesting NYTimes article on the misuse of malaria nets:

    What are GiveWell’s thoughts on this?

Comments are closed.