The GiveWell Blog

March 2018 open thread

Our goal with hosting quarterly open threads 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 or to request a call with GiveWell staff if you have feedback or questions you’d prefer to discuss privately. We’ll try to respond promptly to questions or comments.

You can view our December 2017 open thread here.


  • Are there any really small charities you can recommend, where a gift of around $1,000 would be enough to impact on what the charity does?

  • Eric Anderson on March 14, 2018 at 1:17 pm said:

    Hello there! I was looking over your cause area report on salt iodization, and noticed that while you look into the effect of prenatal iodine on cretinism, and of childhood iodine on cognitive ability, you don’t show any analysis of prenatal iodine on cognitive ability. Some studies (linked below) show a strong effect in areas with severe deficiency.

    Is there a reason you don’t go over that possibility? Either the evidence base not being strong enough, or severe deficiency not being typical of where most programs are able to operate?


  • Cameron Holloway on March 14, 2018 at 3:47 pm said:

    Have GiveWell (and EA funds) considered a ranking system which ranks the most effective charities by cause, as Giving What We Can has? Currently, I worry about the scalability of their approach. Say, as EA becomes more influential, a very large percentage of charitable donations are given to AMF, and they are able to provide bed nets to everybody in the developing world who needs them. This sounds like a good thing, and those funds could then be redirected to the next most effective charity not working on malaria, say Schistosomiasis Control Initiative. We then solve Schistosomiasis, and those funds are redirected again to Give Directly, and in this way we solve the world’s problems one by one. Sounds great, doesn’t it?

    But it just doesn’t work like that. The world’s problems cannot all be dealt with one by one – in fact, pretty much the opposite is true. If we all spend our donations on buying enough bed nets for every person threatened by malaria, that means that we are not providing clean water in Somalia, or food in South Sudan, or doctors in Yemen. The idea, and for now the reality, is that these things will happen anyway, because there are many large organisations focused on disaster relief, providing clean water, education, food etc. and these sorts of projects are more likely to draw donations anyway, as they attract more sympathy. But what if that were no longer the case, as effective altruists must surely desire? Are GiveWell currently looking at trying to cover a wider range of areas with their top charities, and are they concerned about the bias inherent in the fact that some issues, such as malaria and worms, are clearly far more effective to work on than others?

  • Milan Griffes on March 14, 2018 at 3:50 pm said:

    How has Chris Blattman’s 10-year follow-up of a cash transfer program in Uganda affected GiveWell’s view of the effectiveness of GiveDirectly?

    From Blattman’s conversation with Tyler Cowen, the upshot is that non-recipients caught up with transfer recipients after 10 years:

    “We recently went back to some cash transfers that were given almost 10 years ago, following up a randomized control trial in Uganda in the north, and we’re just, in some sense, putting out those results.

    What we found is, the initial result after two and four years was like other places seeing big advances in incomes. People get cash. They’re poor. They couldn’t invest in some of their ideas, but they had good ideas, and so they take off.

    Now what we’ve seen is, essentially, they’ve converged with the people who didn’t get the cash. The people who didn’t get the cash have caught up because they saved and accumulated slowly and got up to the point where they have the same levels of success.

    They converged to a good level. But this means that cash transfers are much more of a temporary acceleration than they are some sort of permanent solution to poverty.”

  • Has GiveWell disbursed any of the finds collected due to their end of 2017 recommendations? Or, has GiveWell released publicly the total amount raised over the key giving season? (I haven’t seen this information, but I may have missed it) And how does the amount compare to the impact of recommendations in previous years?

  • Eric Anderson on March 16, 2018 at 10:59 am said:

    Gary, the best advice I’m aware of on that front is what Peter Hurford put together on the EA forum:

  • Eric Anderson on March 16, 2018 at 11:51 am said:

    Gary, it also might be worth looking into organizations EA Funds have given small grants to, as they try not to be the sole source of funding for any organization, so there are likely to still be some funding gaps remaining.

  • Catherine (GiveWell) on March 21, 2018 at 3:07 pm said:

    Hi gary,

    Though we generally recommend “large” charities (with annual budgets of $1 million or more, due to the amount of funding we expect to direct as a result of our recommendation), we believe $1,000 has an impact, even when given to a very large charity, by allowing them to expand their program.

    You could think about the rationale in a couple different ways:

    (1) If you behave in the way that you want everyone to behave. If everyone said “my $1,000 doesn’t matter,” the charity would get no funding.

    (2) If your $1,000 is the funding that pushes the charity over the tipping point to conduct its work, in the case of groups that move large amounts of funding out the door at once. For example, the Against Malaria Foundation (AMF) operates by signing very large net distribution agreements (see here). If AMF only moves money out the door in $1 million increments (for example), then the first $999,000 donated doesn’t have an impact. However, if you give the last $1,000, you have a huge impact; your $1,000 caused the distribution to happen. So, there might be a low probability (1/1,000) of your $1,000 being the one that causes the distribution to happen, but the impact is so outsized (an extra $1 million distribution) that you may consider that a bet worth taking. The expected marginal value of your donation is therefore (1/1000) x $1m = $1,000 extra spent on net distribution.

    I should also note that AMF links every donation to a specific distribution, so you can see how your donation was used.

    In practice, when there are a large number of donors cooperating, it’s not feasible to identify which donor pushed AMF over a particular threshold. We therefore think it’s reasonable to think of the marginal impact of each donor as scaling linearly with the amount donated. If you’re interested in reading more about the theoretical framework behind donor coordination, you might want to read about Shapley values.

    The case for impact for a $1,000 donation to GiveDirectly is more straightforward; we believe they generally scale up more linearly (as opposed to primarily moving funding out the door in large segments, as in the previous example), such that additional funding is likely to enable them to directly grow their cash transfer program.

  • Catherine (GiveWell) on March 21, 2018 at 3:16 pm said:

    Hi Eric,

    Thanks for passing this along. We haven’t seen this particular paper; we’ll put it on our list of studies to consider as part of updates to our iodine report. From a quick glance, it looks like this paper may be more complex to evaluate due to weaker study designs (observational studies) than what we’re relying on as the core evidence in our current report.

  • Catherine (GiveWell) on March 21, 2018 at 3:18 pm said:

    Hi Milan,

    We have seen Chris Blattman’s interview with Tyler Cowen; however, we have not seen the results yet from the 10-year follow-up he mentions. We look forward to reviewing them.

    As a general rule, we do a full, formal review of new evidence relevant to our priority programs each year. This year, we aim to conduct that review in June and July.

  • Catherine (GiveWell) on March 22, 2018 at 1:39 pm said:

    Hi Cameron,

    Thanks for your questions! First, just a bit of history for context. When GiveWell began, we had a menu of causes for donors to choose from, and recommended what we believed to be the best option in each of those causes (see, for example here). However, we found that our donors were primarily interested in giving to the opportunity we believed to be the most cost-effective overall. We now focus on finding opportunities that we believe to be the most evidence-backed and cost-effective across causes, rather than in each cause.

    We estimate that GiveWell’s top charities received approximately $120 million as a result of our recommendation in 2017 (we’re still finalizing our money moved figures, so this figure may change, but we don’t expect it to change much). We also estimate that our top charities can absorb hundreds of millions of dollars effectively—they have hundreds of millions of dollars of “room for more funding.”

    Once our top charities’ room for more funding is filled, we would recommend people give to other opportunities. In a world where we’re directing all of the donor dollars, the most cost-effective gaps would be filled very quickly, and I expect we would move down the line to direct funding to different causes quite quickly, such that it’s unlikely we would solely be funding malaria, deworming, etc., for very long.

  • Catherine (GiveWell) on March 22, 2018 at 1:41 pm said:

    Hi Matt,

    We decided to grant funds donated for the support of “Grants to recommended charities at GiveWell’s discretion” during Q4 2017 (~$5.6 million) to the Schistosomiasis Control Initiative; a write-up of our rationale is forthcoming on the blog. The funding was disbursed last week. A list of past recipients of discretionary funds can be found here.

    You can see our most recent report on money moved to our top charities at the top of this page; past reports are linked below. We estimate that in our 2017 metrics year (February 2017 through January 2018), we moved about $120 million total to our top charities; this is ~30% higher than our estimate of $91.6 million in 2016. Most of this difference is due to Good Ventures increasing the amount it gave to our recommended charities from $50 million in 2016 to $75 million in 2017.

  • Catherine and Eric,
    Thank you very much for your answers to my question. I will be guided by your answers in my charitable

Comments are closed.