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 info@givewell.org 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 March 2020 open thread here.
Comments
Hi! Thanks for opening up for questions. Assuming it is the latter that GiveWell is aiming to maximize, why does GiveWell believe that DALYs approximate goodness? Relatedly, why does GiveWell believe that DALYs approximate goodness substantially more than consumption and subjective well-being?
I have one more question. Open Phil, in addition to relatively risk-averse giving (e.g. to GiveWell charities), does hits-based giving: bets on non-profits, organizations, and companies with a low probability of producing an outsized impact. The underlying reason is that the overall good produced by hits-based giving may vastly exceed the overall good produced by highly risk-averse giving. If this is true, then wouldn’t GiveWell produce more impact by offering its large and growing grassroots donor-base the opportunity to do the same? I wonder if this offers a solution to the bind which has faced GiveWell so far with respect to things like healthcare ecosystem building, where there is obvious potential but low amenability to GiveWell’s preferred evidence (e.g. RCTs). Even more speculatively, I wonder if investment in human rights activism (e.g. Amnesty International) or grassroots anti-corruption work (e.g. BudgIT in Nigeria) might produce vastly more good than AMF or SMC. I’d ideally like to donate maybe 5-10% of my total donations to these more speculative causes and I’d love for GiveWell to help me speculate wisely.
Hi Parth,
Thanks for your questions!
In response to your first question: While we used to use DALYs in our cost-effectiveness analyses, we no longer do. The decision to stop using DALYs is discussed here (item #2).
GiveWell’s current recommendations focus on improving well-being through averting death and increasing incomes and consumption. Three of the top charities we recommend today support programs that avert deaths (Against Malaria Foundation, Malaria Consortium’s seasonal malaria chemoprevention program, and Helen Keller International’s vitamin A supplementation program) and five support programs that increase incomes and consumption (the four deworming charities we recommend and GiveDirectly). You can see our most up-to-date cost-effectiveness analysis comparing programs here.
In response to your second question: GiveWell operates in an expected value framework that weighs the potential costs of charities’ work against the potential benefits. We’re open to supporting opportunities that might have a very large impact, or might have a much more limited impact, if the expected value is high; this blog post looks at our recommendation of deworming charities through this lens.
In a February 2019 blog post, we wrote how our research is evolving to look at new areas that might impact the lives of people in poor countries. This research would consider a broader range of evidence than used in our traditional work. We have not made charity recommendations based on this work, as it remains in the early stages, but we may in the future.
Has GiveWell done any research into education and childcare (from low income families or high risk backgrounds) based causes/charities from a consumption increase stand point? I’ve historically donated a certain proportion of my donations to the country chapter of SOS Children’s Villages which does work in this area. But after starting to think about giving from an EA point of view, I’ve been trying to revaluate the effectiveness of this. Research in this area has been hard to come by though.
Since you are now factoring in subjective well-being, have you considered looking into the quality of life improvements that could be had by reducing overtime in East Asia (eg. Japan, China). In many countries in that region, children and youth are in class or doing homework all hours of the day and white collar adults generally work constantly, as well. Changing public policy or culture would result in a huge increase in quality of life for hundreds of millions of people. (Also it would save lives: many East Asians die young from extreme overwork). It is extremely high scale and completely neglected far as I can tell. I think it could be tractable if the movement was sufficiently funded.
Hi. I have some internal conflict. On the one hand I would like to travel and explore the world and on the other hand I feel like if I don’t keep working I’m forgoing the opportunity to make money to donate to those who are suffering. I was wondering if there are places in the world to travel to where I might see charity going on first hand, and this would satiate my wanderlust but perhaps more importantly make me *feel* the impact of my donations (currently they are just numbers in a bank account so hard to sacrifice for).
I realize this may not be the right place to ask this question but don’t know where to ask (maybe you even know a better place to ask, if you don’t have a direct answer). Appreciate it a lot 🙂 (also an informal answer of the first thing that comes to mind is plenty and I appreciate it)
Hi Joseph,
We’ve written about our views on education in developing countries as a potential priority program here. We currently believe the evidence for educational programs is less strong than the evidence supporting programs we recommend.
Hi Austen,
Currently, we do not explicitly incorporate charities’ effects on subjective well-being into our assessment of charities’ cost-effectiveness. We assess charities based on their effects on different outcomes, such as deaths averted or increases in consumption, which we translate into a common metric using “moral weights” that assign different values to one death averted or doubling consumption for one person for one year. These moral weights are not generally based on the effects of increasing income or averting death on subjective well-being measures.
We plan to consider the role that subjective well-being measures could play in our work in the future. Thanks for sharing your thoughts!
Hi Maxwell,
Thanks for reaching out! Unfortunately, we aren’t sure which charities have site visit or volunteer programs, and we are not able to arrange potential visit or volunteer opportunities with our top charities.
We’re sorry to not be more helpful with this!
Hi! Can you please suggest a charity for the urgent crisis in Yemen?
Hi Erin,
Thanks for the clarification. I was thinking that you were considering subjective well-being. Overwork is also a major cause of early death — I’ve read a million/year in China and 150k/year in USA, for instance. It is a public health problem that is completely and utterly neglected. But it is tractable. South Korea was the most overworked country and a couple years ago they passed a law that slashed the hours employers could force their employees to work. It’s a taboo space because people scared they won’t be seen as a hard worker if they stand up for work-life balance. Even unions are too scared to go there.
Elie H conveyed (roughly) in a podcast interview last year how buying a cup of coffee for a needy person at your local Starbucks can offer more fulfillment to the donor than making an anonymous online donation of $3k to Against Malaria. He mentioned how generating more meaning for the GiveWell donor experience is needed. I’m curious if any efforts are underway at GiveWell to address this – to add more fulfillment to the donor experience. If so, what is GiveWell considering or working on. Thanks! Matt
Hello,
I would love to know of some good charities for the crisis in Yemen?
Like Emma and Mana, I would also like to know which charities you would recommend for giving to Yemen. Related, I am aware that usually the advice is not necessarily to give to disaster relief/major crises given the outsized media coverage they receive, but wondered whether the logic changes in this case, given that media coverage of the Yemen crisis appears to be minimal currently (likely swamped by coverage of COVID and BLM)?
Hi Mana, Emma and Matt,
Unfortunately, we don’t have recommendations for organizations serving Yemen. We have written some general tips for giving to humanitarian crises in the past; these are available here.
In general, GiveWell does not aim to rate every charity, but to find the ones that are evidence-backed, cost-effective, transparent, and underfunded. For our work identifying potential new top charities, we are currently prioritizing organizations that focus on what we consider to be priority programs in global health and poverty alleviation. To read more about our top charity criteria, see here.
Sorry not to be more helpful with this!
Hi Matt,
Thanks for your question! Yes, we are planning to share more information to better connect donors with the work of our top charities and the impact donations have. This includes updated web content, blog posts such as “GiveWell donors supported more than direct delivery: AMF and new net research” and “Why ongoing assessment of top charities leads to more impact: HKI’s vitamin A supplementation program,” emails to donors on the impact of their gifts and information on the program they’ve supported, and other outreach and marketing efforts.
Has there been any updated evidence regarding deworming?
I’m interested in donating to these programs because they seem extremely cost-effective relative to Malaria Consortium’s SMC program or even HKI’s VAS program. However, I’m weary after reading one of your blog posts about it here: https://blog.givewell.org/2016/07/26/deworming-might-huge-impact-might-close-zero-impact/
GiveWell’s cost-effectiveness analysis assumes that welfare is logarithmic with consumption. This is a fairly common assumption, but some evidence suggests that welfare is sub-logarithmic. For example, in Drupp et al., “Discounting Disentangled” a survey of economists found an average estimated elasticity of marginal utility of 1.35 and a median of 1.0 (an elasticity of 1.0 corresponds to logarithmic utility). Does GiveWell have some justification for assuming logarithmic utility?
Hi Wenmar,
We continue to recommend deworming on the basis that it may have a significant impact on children’s incomes as adults. Because it is so cheap and the potential benefit so large, we think the expected value of deworming is quite high, even though there is a chance that deworming does not have a significant impact on later-in-life incomes. In short, there have not been major updates in our deworming recommendation since the post you linked was published.
You’re correct that in viewing our current cost-effectiveness model, our deworming charities have high overall cost-effectiveness. However, modeled cost-effectiveness is not the only factor we use to prioritize among our top charities. For example, when we make our quarterly allocations of discretionary funding, we also consider the qualitative strength of organizations and what we call ‘room for more funding’, in which we assess the forward funding gaps at each top charity to determine which gap is most pressing and/or which charity could put additional funding to use most productively. In January, we described why Evidence Action’s Deworm the World Initiative, a deworming charity, wasn’t our highest priority funding need on the margin, despite its high cost-effectiveness; you can read more here.
Hi Michael,
You’re right that GiveWell generally uses a logarithmic model of consumption (with an assumed elasticity of marginal utility of consumption of 1). In other words, we value increasing someone’s consumption by X% as equally valuable regardless of their starting level. When we first included this estimate in our cost-effectiveness model, it seemed like a reasonable simplifying assumption.
We have not conducted a thorough review of the literature on the elasticity of marginal utility of consumption, nor fully evaluated arguments for whether values other than 1 may be appropriate. Thank you for raising this issue—we may spend more time investigating this topic in the future.
Hi Givewell! I’m a big fan!
I don’t know if this the right place for this question, but can’t hurt to ask. I am interested in doing research in the nonprofit sector and I would like to study how nonprofits compete for donations. I have been looking at IRS Form 990 data, but I don’t think any line item corresponds very well to how much effort or investment a nonprofit dedicates to fundraising or soliciting donations. Ideally I would like to have something like the total cost of fundraising, including wages of personnel directly involved, etc. Are you aware of any such data? Any help would be much appreciated!
Hi JMG,
Unfortunately, evaluating nonprofit fundraising like you’ve described is beyond the scope of our research. I’m sorry we can’t help!
It seems that the discresionary grantmaking page has not yet been updated for the June 2020 discresionary grant. Do you know when that information will be available?
Thanks!
Hi Luke,
We have not yet updated our discretionary grantmaking page because we delayed allocating donations made to our discretionary fund in the first quarter of 2020. Due to the unusual amount of uncertainty introduced by the COVID-19 pandemic, we wanted to gain a clearer understanding of how the pandemic will affect the plans, budgets, and operations of our recommended charities. We believe this delay will enable us to better understand how donations can have the greatest impact and to ensure they’re allocated according to where we see the greatest need. We believe the benefits of delaying outweigh those of getting the funding to charities sooner.
We emailed donors who gave to our discretionary fund in the first quarter to let them know of the delay in May. We plan on updating our website and informing our donors once we allocate discretionary funding from the first and second quarters of 2020.
Hi Erin,
Thanks for opening up for questions. Assuming it is the latter that GiveWell is aiming to maximize, has GiveWell done any research into education and childcare (from low-income families or high risk backgrounds) based causes/charities from a consumption increase stand point?
Hi ofhsoupkitchen,
We don’t currently recommend any programs focused on education or childcare. We’ve written about our views on education here; we currently think the evidence for educational programs is less strong than the evidence supporting programs we recommend.
I am gratefully informed by the thoughtful cost-effective analyses. Estimating that deworming improves life a lot opened my eyes. Likewise, I found some answers to my curiosity on how much education improves life. I also felt informed to adjust the moral weights.
Does a reader here know how the effective education spreadsheet could be adjusted to account for an outlook that values the average quality of life for the next generation? One of the values of girls education is the improvement of life for the next generation due to finances, health, and family planning. In my imagination, an educated woman has a compounding benefit for every generation that follows. I saw a cell in the spreadsheet for the value of one fewer child. Yet modifications to that cell seemed have at most some kind of influence that was not compounding, rather it appeared to dampen whatever number I entered. For example, I changed the staff’s optimistic value (Parameters cell B13) of 0.2 to 5. The outcome only changed the cost-effectiveness ratio (DDK 2017 cell B94) from 2.6 to 3.0.
For the time being I accept that my outlook on the most effective educational programs is only 3x as effective as Give Directly. And that deworming is 20x as effective as the best education. I recognize evidence that deworming may be one cause of a long-term benefit. Such as the recent 20-year follow-up.
Yet it seems like there’s a reasonable model that would show at least a little compounding of benefits from an employed and planning mother. The children of that mother live better, and in turn may raise grandchildren that live exponentially better. In part from focusing love and attention more deeply on fewer children. In this model, the quality of the children, not the quantity of children, is the ideal outcome.
Hi Ethan,
Thank you for your engagement with this topic!
It appears that you’ve found the page on our website where we discuss our views on education; we currently think the evidence for educational programs is less strong than the evidence supporting programs we recommend. We have not prioritized additional work in this area in recent years. As such, we haven’t updated our cost-effectiveness analysis for education in a few years, so that content may not be fully up-to-date.
As an organization, we see our strength in identifying opportunities to help people in low- and middle-income countries over relatively short time horizons, and with relatively high levels of confidence.
We limit the quantitative outputs in our cost-effectiveness model to things we believe we can reasonably quantify, and we don’t believe we can make useful estimates about outcomes in the distant future. We do think that some things that we can’t robustly quantify are important, but we consider those outside of our model.
Thanks again!
Hi GiveWell team! Thank you for everything that you do. I’m seeking some advice on how I can donate beyond just my money. I understand the monetary funds are very important, but to be so removed from the process makes me feel not very involved with the charitable cause. I have considered tutoring under-represented teens in my area of expertise but the issue of effectiveness always lingers. Suppose I tutor a child for 10 hours a week, if I put that 10 hours towards my career that increases my salary by 10K of which I donate, that’s a lot more effective than helping one person. What are you thoughts? Thank you!
Hi Ryan,
Thanks for reaching out! Unfortunately, I don’t have any recommendations for you, as GiveWell’s mission is to direct donor funding to the best giving opportunities for saving or improving lives that we can find. We haven’t focused on other types of aid, such as volunteering.
In case it’s helpful, you might be interested in this post from GiveWell Board member Julia Wise, which discusses how she thinks about similar decisions.
Comments are closed.