The GiveWell Blog

Staff members’ personal donations

For this post, GiveWell staff members wrote up the thinking behind their personal donations for the year.

Holden Karnofsky

There are good arguments that I shouldn’t give anything this year. I tend to think that future giving opportunities will be better than this year’s, even accounting for the costs of delaying. I expect the best giving opportunities I’m aware of (according to my views) to be funded by Good Ventures or other donors. The best giving opportunity I see that Good Ventures can’t fill is diversifying GiveWell’s donor base, but it doesn’t make much sense for me to donate to GiveWell (it would make more sense to simply lower my salary, or save money so that I can lower my salary later if need be).

Yet it wouldn’t feel right to skip giving to charity altogether (or to give to the organization that employs me). In general, I try to behave as I would like others to behave: I try to perform very well on “standard” generosity and ethics, and overlay my more personal, debatable, potentially-biased agenda on top of that rather than in replacement of it. I wouldn’t steal money to give it to our top charities; I wouldn’t skip an important family event (even one that had little meaning for me) in order to save time for GiveWell work; and along the same lines, I want to give some substantial portion of my money each year in a way that (a) aims at helping the less fortunate (b) can’t be confused with aiming to benefit myself or those in my circle. In addition, cutting a check for the sole purpose of helping others is something I want to stay practiced at.

(Nothing in the above paragraph is absolute. Under certain circumstances, I could imagine reserving my entire charitable budget for a friend’s project; I could also imagine skipping an important family event if the urgency for GiveWell were great enough. But the statements in the above paragraph do represent values that carry substantial weight for me.)

In choosing between our top charities, the main question on my mind is how to directly improve lives the most for my money. (This contrasts with last year, when we asked donors to follow a target allocation in order to optimize incentives, learning opportunities, and capital for organization building. This year, we’ve accomplished the same thing by setting minimum targets, and thanks to Good Ventures, we expect these minimum targets to be hit regardless of what people like myself do.) Because DtWI has a relatively small funding gap that I expect to be closed, I see this year’s choice as being between GiveDirectly and SCI.

In my mind, the question of which has higher expected direct-life-improvement-per-dollar is a difficult one.

  • I have more confidence in the impact of a gift to GiveDirectly. I understand the organization and its process better, and I understand the intervention better. I don’t have the negative intuitive reactions that many have to cash transfers: to me it just feels like the simplest, most efficient way to transfer my wealth to people with less. By contrast, I think of deworming as an intervention whose main case rests on two problematic studies. To compensate for the difference in confidence, the difference in cost-effectiveness ought to be substantial.
  • On the other hand, our best guess says that deworming does about 3x as much good per dollar spent, even accounting as neutrally as we can for the limitations to and questions around the evidence. This is a highly non-robust calculation (see Elie’s comment on how widely reasonable estimates can range), and all else equal I expect further corrections to move me in a more pessimistic direction about deworming (as has been the general pattern with our cost-effectiveness estimates). But 3x – after accounting for concerns over study validity – is a pretty substantial multiplier, potentially enough to overcome my misgivings.

If the multiplier were 10x, I’d feel confident that SCI wins on this dimension. If it were 1.5x, I’d feel confident that GiveDirectly wins on this dimension. But I’m not sure what to do with 3x. The question here largely seems to be how much to discount explicit expected-value estimates, a judgment that I don’t know of a good way to formalize or vet. One of the things I’ve done as I’ve thought through this is talked multiple friends through the issues and – after I’m confident they understand the tradeoffs – asked what they would do. The answers have been mixed, and all of the people I’ve consulted have considered the question of which has the higher expected direct-good-accomplished-per-dollar to be a tough one with 3x being in the “gray area’ mentioned above.

I think that if I were exclusively considering this dimension, I’d be leaning toward SCI, but it would be a very difficult choice. Adding in even a small weight for some of the less tangible factors (which remain factors, though less so than last year) leaves me fairly comfortable with a choice of supporting GiveDirectly.

Elie Hassenfeld

I agree with everything Holden wrote except for his perspective on cost-effectiveness. In the case of deworming, I think there are too many unknowns to put weight on our “best guess” and we’re better served not relying on the results of our cost-effectiveness model when trying to estimate the impact of deworming.

I agree with Holden’s arguments for waiting. The reason I’m giving this year is that given that we’ve identified ways to significantly improve the world via our top charities, something seems wrong about holding my money instead of giving it. I recognize that, on expectation, giving now doesn’t maximize utility, but maintaining a commitment to give substantially every year is important to me. I’m supporting GiveDirectly with the full amount of my charitable donation, but planning to ask GiveDirectly to not include my funds in those counted towards the Good Ventures match (for reasons discussed here).

Alexander Berger

I expect there to be somewhat better giving opportunities next year, but I also find both personal and rhetorical value in giving significantly at least once a year, so I’ve decided to split my donation between GiveDirectly and waiting. I’m planning to give the amount to GiveDirectly that I typically suggest other people (who are able to) should aim for, and save the rest of what I would otherwise give for donating in the future.

Even though I don’t think it necessarily maximizes the return on my charitable dollars–though I’m also not sure it doesn’t–I think this is a good decision for two reasons:

  1. I believe that GiveDirectly is a more than good enough giving opportunity to get the moral case for giving (a lot) off the ground, and I believe that actually giving, rather than saying that I save to give in the future, will make me a more effective advocate for this view when talking to others.
  2. Separately, as a GiveWell employee, I think that when I’m talking to people about my work it’s valuable to be able to say that I use our research. I also find the idea of “eating [my] own dog food” fairly important for myself.

I didn’t find the choice between GiveDirectly and SCI as difficult as Holden did. While I acknowledge the higher estimated cost-effectiveness of SCI, I agree with Elie’s criticism of the limited value of such estimates (though I don’t think they should be fully discarded). I continue to see donations to GiveDirectly as carrying high upside, by helping accelerate the growth of a new organization and the dissemination of the principle that cash is the appropriate benchmark for other charitable interventions, and I find that to be easily enough to outweigh the higher estimated cost-effectiveness for SCI.

That said, these “upside” considerations also present two significant risks that I don’t think we’ve necessarily discussed at great length elsewhere (partly, I believe, because other GiveWell staff are less motivated by the upside consideration than I am):

  1. GiveDirectly’s growth is transient rather than sustained. Most GiveWell donors change their allocations when our recommendations change, and I would expect us to recommend GiveDirectly less highly next year than we do this year (though of course things could change). If GiveDirectly does not develop a significant, sustainable, non-GiveWell funding stream, I think there’s a real risk that this year represents a one-year blip rather than an overall acceleration in their growth as an organization. Accordingly, I’d be excited to see them try to become an organization that would be more successful with fundraising from ordinary individual donors; I also think such donors make up a more natural target for them, since in the long run I expect foundations and GiveWell-driven individual donors to have opportunities that do beat cash. That said, their plan to provide services to government cash transfer programs also strikes me as a reasonable attempt to address this issue.
  2. GiveDirectly focuses on “just moving money to people.” Because so much of the upside I see in GiveDirectly comes from helping establish the idea of cash as a baseline, I think it would be a problem if they focus too narrowly on scaling their existing model, rather than conducting further research and experimentation. (One question I would be excited to see GiveDirectly explore more rigorously is the best size for transfers; I wonder if the current transfers might be large enough that declining marginal returns significantly curtail their value.) I worry that in the course of trying to significantly expand the organization, GiveDirectly might focus too narrowly on simply repeating what they have done before. That said, they have explicit plans for further experimentation, which appear to go a long way towards addressing this concern.

Overall, I find GiveDirectly an outstanding giving opportunity for this year, and I’m excited to donate there.

Natalie Crispin

I feel strongly about giving each year. While my charitable giving is quite modest in the scheme of things, it is generally one of, if not the, largest single expenditure that I make in a year and, as with any large expenditure, I have felt trigger-shy when it comes time to write the check. That’s why, in part, it’s important to me to build up the habit of giving substantially and to work to be more generous as time goes by. I find this couple’s plan to give a little more each year an inspiring way to approach giving.

I plan to support GiveDirectly this year. I considered splitting my gift between GiveDirectly and SCI, but ultimately decided that I have more confidence in GiveDirectly and I believe that SCI will receive substantial funding as a result of our recommendation. I decided not to donate to Deworm the World Initiative for the reason Holden noted.

Eliza Scheffler

As someone who is early in her career and therefore early in her charitable giving, it’s important to me that I get into the practice of giving generously and effectively. In order to formalize this practice, I pre-committed at the beginning of the year to donate a percentage of my income to the organization(s) that I feel will maximize the impact of marginal donations. This year, I’ve decided to give the full amount of my pre-committed charitable giving to GiveDirectly.

Of any of our top charities, I have the greatest confidence in GiveDirectly’s impact and ability to scale effectively. Given that I expect our minimum fundraising targets to be met, I feel best about investing additional dollars in GiveDirectly’s long-term growth. Part of my view is based on our impression that most proven, cost-effective interventions attract substantial funding from governments and international institutions. For this reason, I expect (and hope) that the funding gap for deworming is likely to decrease significantly in the not too distant future, and therefore the room for more funding for deworming charities could be substantially limited. However, the funding gap for cash transfers seems to me, for all intents and purposes, unbounded. I want to invest in GiveDirectly because I feel that it could continue to use funds well as a significantly larger organization in the near and farther future. I also feel there are benefits to building the movement around cash transfers for anti-poverty work more broadly.

I decided to make a much smaller donation this year to Mercy for Animals (MFA). Neither GiveWell nor I have conducted an in-depth evaluation of MFA or a systematic scan of farm animal charities at this point. (For example, I have not looked into Compassion Over Killing, which appears to play a somewhat similar role to MFA. I plan to do more research on this field before next year’s giving season). My impressions of MFA are based mainly off of one conversation I had with Nathan Runkle, its Executive Director (notes forthcoming), and therefore my knowledge and confidence are limited. I decided to support MFA because it works exclusively on farm animal issues, which seem disproportionately underfunded, and because it conducts undercover investigations, which I preliminarily believe to be an important and potentially high-leverage tool for farm animal advocacy. I’m giving to MFA now, rather than waiting for GiveWell to conduct more research in this area, because I expect multiple ag gag bills to be introduced this year and feel that MFA may be particularly well-suited to challenge them. MFA is also a relatively young organization, and I believe there could be benefits to funding it early.

Because of my relative lack of knowledge about MFA or the likely impacts of my donation, I didn’t feel comfortable removing this amount from my pre-committed funds for charitable giving. Instead, I’m taking this donation out of the money I would otherwise have spent on personal consumption or savings. There are other very small donations that I plan to make this year that I include in the same category of “alternative personal consumption,” because they are not aimed at maximizing impact, and they are all personally beneficial or meaningful to me to some degree. These funds include membership fees for civic organizations, money I use to purchase food for people I meet who appear financially impoverished and ask me for food, contributions to my friends’ and family members’ projects, and donations to support the religious community groups in which I participate.

Timothy Telleen-Lawton

Personal giving is a decision with many plausible decision methodologies that could result in very different answers. When facing such a decision, I think it’s useful to entertain several methodologies simultaneously and compare the results with each other and with my intuition rather than focus on making the best single chain of logic as strong as it can be and then accept whatever answer that produces. (More on this idea here.) Full reliance on Cost Effectiveness Analyses is one of the clearest examples of such “best single chain” reasoning, which I think is unnecessarily risky.

My process resulted in an allocation that may be best understood by the following driving factors:

  • Don’t let the publication of my decision or my employment with GiveWell affect my giving decision (for example, don’t avoid giving to GiveWell just because this would make me appear self-serving). This is because I think one of the best values of this post is to give people a chance to see how GiveWell staff think ‘behind the scenes’.
  • When there are clear opportunities to do good, err on the side of giving rather than saving. In addition to the reasons to do this that Holden and Elie mentioned above, I am also not as confident as Elie that giving rather than saving “doesn’t maximize utility” since I have more confidence in GiveDirectly in 2013 than the theoretical opportunities in 2014 (from today’s perspective).
  • When publicly available information supporting top charities is weaker, giving opportunities based on private information becomes relatively more compelling. (See the bottom of this post.)
  • Look for opportunities to engage others in the process of giving thoughtfully in a scalable way (ie not based on private information or overanalyzing how to interact with recommended top charities).
  • If I want to give to something that doesn’t have good impact-maximizing justification, consider its value as miscellaneous spending for personal happiness (and allocate the gift size accordingly), rather than as weakly justified altruism.

Thus I am splitting my charitable gifts into three categories of unequal size:

  1. Personal connection/service giving: I make token recurring gifts to a few non-profits that I have a personal connection to or am served by. I believe each of these has a positive social impact, but I’m not at all sure that this spending is well-justified as effective altruism.
  2. Engagement giving (donations in the name of friends and family members for Christmas or birthdays): All my engagement gifts will be to GiveDirectly or Evidence Action depending on the recipient. Recipients that are excited about traditional aid projects or would be compelled by cost effectiveness estimates will get Evidence Action. Those that are skeptical of traditional aid projects or non-profits generally, think of giving as consumption, want to be as-confident-as-possible in the impact of their gift, are excited by utilizing market forces for better decision-making, or don’t clearly fall into any of these categories will get GiveDirectly. The main reasons I won’t be giving SCI gifts in place of Evidence Action is my excitement for Evidence Action’s evidence-based vision, their staff which I have met, and the organization’s relative youth (such that it may be more influenced by marginal funding than a more mature organization), as well as the SCI update described in this post. My best guess values for our cost effectiveness analysis imply that SCI’s program is slightly more cost-effective than Evidence Action’s program, but I don’t put a lot of weight in that analysis, especially for differences of less than a factor of 2.
  3. Opportunistic giving: If my decision were anonymous I would also give significantly to GiveWell unrestricted this year (and thus am doing so), largely for reasons that I expect to be less true in future years. While it hasn’t received an independent GiveWell-style review, I am nonetheless highly confident in its value for improving the quality of information available for givers in the future and its need for funding now (assuming Good Ventures funding is or should be capped). Critically, I don’t expect it to hit its “minimum funding target”, unlike my expectation for the top charities. Even if GiveWell would ask Good Ventures to break the cap rather than cut its own spending (thus implying that donations to GiveWell are partially fungible with Good Ventures money) I don’t mind being partially funged by Good Ventures because I consider the foundation to be “smart money” and I think it’s appropriate for funders that agree on a good giving opportunity to share the burden of that cost rather than play a competitive “funging game”.

Sean Conley

This year I’m planning to split my giving between GiveDirectly and SCI.

I’m planning to give this year (rather than save for future donations) because I believe that our current top charities represent outstanding giving opportunities and view the potential for better opportunities in the future as uncertain.

Cash transfers and deworming are both programs that are likely to be highly impactful. There seems to be too much uncertainty for any decision between the two to be particularly robust, so I am choosing to support both.

While the Deworm the Word Initiative is also an exceptional organization, I put relatively less weight on its potential to leverage donations and relatively more weight on the fact that it is likely to work in areas with lower worm prevalence than SCI, so my preference is to donate to SCI.

Josh Rosenberg

I agree with Holden that the key questions for GiveWell followers this year are whether to give now or give later and whether to support GiveDirectly or SCI (though I also strongly considered giving to Evidence Action).

After much personal deliberation and late-night discussions with other staff members, I’ve decided to give away half of my planned 2013 charitable budget this year and to save half of it to give away during next giving season at the latest. I’ll be supporting GiveDirectly with my donation this year.

Giving now vs. giving later

I’m giving half of my planned 2013 charitable budget now because I think that GiveDirectly is an excellent giving opportunity, I want to build the habit of making large, yearly donations, I want to concretely commit to my beliefs about my obligation to help people who are worse off than me, and I want to be a credible advocate to friends and family considering giving to GiveWell top charities this year.

I’m saving half of my charitable budget until later because I expect that I may learn of more underfunded and higher impact giving opportunities in the near future, especially because it seems plausible that many health-related giving opportunities could “beat” giving away cash. Particular opportunities that I will be keeping an eye on are:

  • Whether AMF is able to finalize a net distribution
  • Whether SCI shares higher quality information next year
  • DtWI’s progress and room for more funding situation
  • The progress of GiveWell’s strategy to find new top (non-Labs) charities next year
  • The progress of GiveWell Labs

Furthermore, the downside of saving seems low. If all goes well with GiveDirectly this year, they may have a larger funding gap and an even more effective model for giving away cash next year.

By (now publicly) committing to give my saved funds by next giving season, I avoid the problem of feeling like I should wait every year until I find the perfect giving opportunity. This commitment makes me accountable to the decision I made this year and gives me a reasonable timeline to reflect on whether I made the right decision.

Finally, since this is my first year out of college (and, consequently, the first time that I’ve had any significant amount of money to my name), my donation this year will be the largest that I’ve made in my life even though I’m saving half of it. I do not expect that I’ll miss out on any of the psychological/commitment benefits by saving some money this year to give away later.

Choosing between the top charities

My case for GiveDirectly includes:

  • Simple intervention with high confidence of impact
  • Trusts beneficiaries to know what is best for themselves
  • Extremely transparent
  • Great learning opportunity
  • High upside because it may help to popularize cash transfers

The reasons that I’m not donating to SCI include:

  • Recent concerns about evidence of impact
  • Has not been a great learning opportunity so far (in my opinion) because of communication difficulties
  • Possible fungibility with major funders like DfID
  • Seems to have relatively low upside

I’m taking the Good Ventures match on my GiveDirectly donation because (a) I don’t expect that taking the match will offset Good Ventures’s giving until further down the line (at least 10 years), (b) all else equal, doing more charity now seems better (because the world is likely to be better off in the future, because of compounding benefits to GiveDirectly recipients, and because of general uncertainty about the future), and (c) I think it is unlikely that the match will be closed fast enough that other donors will miss out on it.

Other giving

Broadly consistent with the reasoning laid out in this GiveWell post, I think it is important to do some “good citizenship”-style giving, though this giving does not come out of my allocated charity budget and is relatively small. As an avid Wikipedia user, a lover of free podcasts, and a person who benefits from many political initiatives in my community, I’ve decided to commit some donations to these causes.

Ben Rachbach

I’m giving half of the total amount I had planned to give to charity this year to GiveDirectly, and saving the rest to give once I have better giving opportunities, probably within a year.

I’m giving this year because I believe that it is important for me to give every year in order to get into the habit of giving. Also, as a GiveWell employee who advocates to friends and family that they should give to GiveWell top charities, it is important for me to put my money where my mouth is and give to GiveWell top charities yearly.

I’m giving to GiveDirectly rather than SCI or DtWI for reasons captured well in other staff member’s entries on this page, especially that I am particularly excited about GiveDirectly as an organization and excited about cash as the benchmark for effective aid.

I will accept Good Ventures’s match for my donation to GiveDirectly. I expect Good Ventures to make even better gifts in the future, but I am uncertain enough about this to prefer that Good Ventures makes a gift to GiveDirectly now rather than holding money for later.

I am saving half of the money I had planned to give this year because I believe that better giving opportunities may soon be available.

First, I believe that GiveWell is likely to find giving opportunities within the next year for which my marginal dollar will have a greater impact than giving to GiveDirectly now.

Second, instead of giving to a charity recommended by GiveWell, I might give to a charity that GiveWell may not investigate soon. I’m particularly interested in giving to a farm animal welfare charity such as Mercy for Animals or the Humane Society of the United States (earmarked for its farm animal welfare work) because I think that this is potentially an excellent giving opportunity according to my values. I may value animal welfare more than GiveWell. A combination of placing a potentially lower value on animal welfare and of strategic choices about which causes to pursue first in GiveWell Labs may lead to a substantial delay in GiveWell or Good Ventures directing funds to farm animal welfare charities, or they may never direct funds to such charities. As such, if I give soon to such a charity, my donation is particularly unlikely to be fungible with funds directed by GiveWell/Good Ventures and thus may have a higher impact. I need to do more research on such charities and have more discussions with GiveWell staff about GiveWell’s plans before I am ready to give, though.

I’ll comment on this post explaining where I chose to give if, as I plan, I end up giving within a year.

Howie Lempel

I believe that this year there has been a temporary dip in the cost-effectiveness of GiveWell’s recommended charities. In particular, I’d guess that cash is substantially less cost effective than the best developing world health interventions. I think this dip will be temporary because:

  • AMF may execute a distribution in the next year.
  • GiveWell’s staff size recently doubled.
  • GiveWell only very recently learned that AMF would not have room for more funding.

The best counterargument I’m aware of is that GiveWell has tried and failed to find more funding gaps for health interventions in the past.

I am persuaded by Carl Shulman’s argument that a temporary dip in recommendation quality favors delaying donations to GiveWell’s top picks, so I ought to hold off on donating. I plan to give half my planned amount anyway because giving is something I look forward to every year. By saving the other half to give in the future, I’ll be breaking a pledge to give away a certain percentage of my income every year.

I’m very excited to split my donations between unrestricted funding for GiveWell and GiveDirectly. I’m giving to GiveWell because I expect supporting GiveWell’s research to have a higher impact than any other giving opportunity I’m aware of.

I also want to give money to an organization that actually implements programs and an organization that is not my employer. I chose to give to GiveDirectly because I think that cash grants make enormous impacts on poor people’s lives while respecting their wisdom and autonomy. I disagree with Good Ventures’ decision to use a match and would prefer that my donation not appear to be influenced by the match, so I’m going to ask that my donation not be matched. For reasons Holden addressed, I also struggled to decide between cash and deworming.

I thought seriously about giving to a couple of other organizations that I thought might be even more cost effective than GiveDirectly. In particular, I thought about giving to a farmed animal welfare charity and have decided not to yet for reasons similar to Ben’s. I also thought seriously about giving unrestricted funds to Evidence Action because I’m excited about their upside but I believe they will already have enough funding to grow a reasonable amount this year and I don’t think they’re an organization that could scale as quickly and easily as GiveDirectly.

Jake Marcus

I feel indebted to my colleagues and to a broader community of effective altruists for their thoughtfulness on how best to give. My discussions with this group have led me to consider the following questions for my giving this year:

  • Should I give now or give later?
  • Should I give now to giving opportunities not recommended by GiveWell?
  • Should I give now to GiveWell?
  • Should I give now to GiveWell’s top charities?
  • How much should I give now vs. give later?

Should I give now or give later?

I’m committed to setting aside a percentage of my income every year for charity. As a GiveWell employee, it helps me think more critically about where I recommend others donate. It also helps me build a habit of giving and hopefully inspires others to give. I think that contributing this set amount to a donor-advised fund largely accomplishes these goals, therefore this commitment does not preclude deferring my choice of where to give in order to take advantage of better giving opportunities in the future.

Should I give now to giving opportunities not recommended by GiveWell?

The large amount of money now moved by GiveWell’s recommendations and the greater involvement of Good Ventures have led some to consider the impact they might have by diverging from this pool of funds, especially if they hold different values on particular issues. I haven’t seriously considered this option, because I feel my values closely align with these two organizations and, as a relative newcomer to philanthropy, still place a lot of trust in expert opinion on where to give.

Should I give now to GiveWell?

Since GiveWell projects a funding gap, some have considered donating to GiveWell itself. I believe that donating to GiveWell represents an outstanding giving opportunity and would strongly consider doing so if I were not an employee. Because I think GiveWell will succeed in closing the funding gap, I’d prefer for others to give thereby diversifying the organization’s donor base rather than to have current stakeholders, such as employees, further increase their commitments. In most cases when talking to more casual donors, I still recommend giving to top charities rather than GiveWell, because I believe an argument that doesn’t seem self-serving will be more persuasive.

Should I give now to GiveWell’s top charities?

I believe that this year’s top charities represent outstanding giving opportunities.
For most potential donors, I think that giving to top charities this year will only increase their involvement in future years. Though the number of donors interested in effective giving has greatly increased over the past few years, effective altruism remains in its infancy. Therefore, I feel that giving to top charities has additional value in so much as it counts as a “vote” for the importance of combating global poverty and promoting transparency in philanthropy. It demonstrates to the world the growing influence and acceptance of these ideas. I also think that giving some amount of money to top charities each year gives me more credibility in persuading others to give than if I save my entire charitable budget to give later. For these reasons, I’m giving to top charities this year.

For the additional dollar going into top charities now, I have the highest expectation for GiveDirectly, but I also feel that a donation to SCI and Evidence Action has some value in casting a vote for top charities. Therefore, I’m giving the bulk of what I give now to GiveDirectly and small donations to SCI and the Deworm the World Initiative.

How much should I give now vs. give later?

I originally planned to give now rather than saving to give later, but then reversed my decision. I realized that I had implicitly lowered the level of my charitable budget, because of my discomfort, but then argued to give “all” of my lowered charitable budget. For me, this reversal underscores the importance of choosing to donate an emotionally salient amount of money if you want it to motivate you to think critically about where to give.

With the exception of the smaller percentage of my charitable budget that I’m giving to top charities in order to accomplish the goals I describe in the previous section, I plan to save to give later by contributing to a donor-advised fund.

As large an impact as giving to top charities now might have, I’m optimistic that GiveWell will find even better giving opportunities in the near future. I’m not sure that the gap between this year’s best giving opportunity and next year’s will greatly exceed the gap between next year’s best giving opportunities and the best giving opportunities 2 years from now and so on for the next several years as the organization continues to rapidly grow and learn. Still, I’m inclined to believe that this year I should save to give later. On a personal level, I have a good reason to believe that this coming year might influence my choice of where to give more than other years, because I recently started as a GiveWell employee and thus will be thinking a lot more about philanthropy.

Process for putting together this post:

  • Holden wrote the intro and his entry and sent to Elie. Elie wrote his entry.
  • Holden asked staff to tell him privately (a) whether they’d like to write an entry; (b) whether they’d like to be anonymous; (c) whether they’d like to see his and Elie’s entries first. All staff chose to write something, preferred not to be anonymous, and chose to see Holden’s and Elie’s entries first.
  • Most staff met a few days ago to discuss their thinking. (Holden and Elie did not attend.)
  • Staff sent their entries privately to Holden, who assembled them in order of seniority (length of time employed full-time), with the exception of himself, Elie and Alexander.
  • The draft post was sent out to all staff yesterday. At that point staff were able to make final edits.


  • Peter Hurford on December 12, 2013 at 6:50 pm said:

    Thanks! This was a very useful compilation. I hope GiveWell continues this annually.

  • Joelle Wyser-Pratte on December 12, 2013 at 6:50 pm said:

    Great post. These are the testimonials of the members of Givewell.. It is worth reading

  • Fascinating read. Did you consider disclosing the amounts or percentages you donated? Many people find it impolite to discuss money, but I wonder if a goal of the effective altruism movent should be to fight this stigma. Perhaps if we were all more open, others would be inclined to give more generously.

    Could also have the effective of appearing elitist, etc., so definitely potential downsides.

  • Ian Turner on December 13, 2013 at 7:48 am said:

    Cody: From my perspective, the GiveWell staff members’ donations are more a matter of signalling than anything else. After all, they are giving up a huge amount of potential income just to be working for GiveWell in the first place, and if their salaries were just a little bit less, then Givewell could grant the money instead, with improved tax consequences. I call it signalling seems to me that paying a cash salary to its employees, then watching them re-gift it away, seems to be designed for tax inefficiency but for good storytelling.

  • Vollmer on December 13, 2013 at 4:56 pm said:

    A question to Eliza: Have you looked at Effective Animal Activism, and what do you think of their research? Is there a particular reason for going for MFA instead of THL?

    … and a question to everyone: Would you give to a meta charity such as Effective Fundraising, and if yes, how would you evaluate such a meta charity?

  • Thanks for the great stuff guys, really thought-provoking. I would definitely support a tradition of supplementing GiveWell’s collective recommendations with individual employee takes.

    Holden posted a few years back about giving now versus giving later, where he offered that “We don’t think there’s a clear answer… [but Holden’s] favored strategy at the moment is to give regularly (a set percentage of my income each year).” (

    The recommendations this year seem to indicate that many disagree with the “fixed % of income” method of giving for a number of reasons elaborated above. If you are confident that the minimum funding thresholds for these charities are likely to be met, have you considered offering additional recommended charities – perhaps including an well researched take on Eliza’s (and Vollmer’s) question of the best animal-related charity – for whom a marginal dollar might go farther once those minimums are met?

    Additionally, what are other’s takes on Alexander’s point that the flow through effects of giving to GiveDirectly, via the “dissemination of the principle that cash is the appropriate benchmark for other charitable interventions” is still more valuable than any other available charity this year, and how the answer changes once the minimum recommendation is achieved?

    Assuming there are people who do not want to wait a year on their donations, it might be of service to offer additional recommendations once all those three minimum funding levels are achieved. Is the potential downside of donors overcompensating their giving in anticipation of the minimums being met towards those hypothetical “second tier” recommendations higher than the benefits to donors whom have already decided how much they are going to give and find the minimums already exceeded?

  • Peter Hurford on December 16, 2013 at 4:43 pm said:

    @Hudson: GiveWell also set targets above the minimum targets which they are comfortable seeing these charities reach (but perhaps not exceed). For example, GiveWell is ok with GiveDirectly getting up to $20M, which is $17.5M more than the minimum target.

  • max hickerson on December 17, 2013 at 11:38 am said:

    You should check out-Operation Finally Home. Started by a retired builder in New Braunfels, TX [near Austin] that builds homes for wounded servicemen and women in many states. Their web site looks very good !!!

  • Thanks for the comments, all.

    Cody: we have considered disclosing amounts/percentage of income, but decided not to do so. We feel that disclosures from some staff members might put implicit pressure on others to disclose, and we feel that many people have valid reasons to keep their giving amounts private.

    Ian, I don’t think anyone’s donations are being made primarily, or even partly, for purposes of storytelling. The setup you propose (in which donations are withheld from the paycheck) isn’t something we’re set up to do or have checked out the regulatory ramifications of, and doing so is not a priority. (It would also provide employees making gifts with less privacy).

    Vollmer, I don’t have a principled objection to giving to meta-charities, and indeed believe GiveWell itself to be an outstanding giving opportunity. I would approach such a gift more with a giving as investment mindset, focusing on the quality of the team, accomplishments to date, my picture of sufficiently likely future possible outcomes, etc.

    Hudson, I think Peter Hurford answered your main question. I don’t think the effect described by Alexander dominates other charities on its own, but it contributes to the case for GiveDirectly (less so, but not zero, as donations to GiveDirectly rise and the minimum target is hit).

  • Vollmer: Thanks for the question! I did take a look at Effective Animal Activism’s recommendations when doing my personal research. I didn’t find enough information available for me to be persuaded by their conclusions. (EAA’s evaluations are very brief and don’t explain the data or reasoning behind some of their main conclusions, about room for more funding, for example). EAA also cites online veg ads as the most cost-effective intervention, but their main source for estimating the impact of these ads is based on self-reported, short-term data on dietary choices, which I’d expect to be highly unreliable (the source acknowledges some of its limitations).

    I do want to be clear that I don’t consider quantitative estimates of impact to be the only important feature of an evaluation. Other factors that I consider include an organization’s track record, leadership, and opinions about the organization from those who are well-informed and active in the field. While I have not done enough research to have a strong view on these factors with regards to Mercy For Animals, I was influenced by a point that Paul Shapiro of the Humane Society of the United States made that MFA is among the groups that the factory farming industry appears to be most concerned about.

    Regarding EAA, I had a conversation with its Executive Director, Jon Bockman, earlier this year to get an overview of the farm animal welfare field as part of GiveWell’s investigation, and I found the conversation helpful. I am looking forward to following EAA as it develops, as outlined on its blog.

  • Vollmer on December 19, 2013 at 3:51 pm said:

    Thanks for the replies!

  • Ian Turner on December 19, 2013 at 10:09 pm said:


    Maybe my remark was not expressed well. I didn’t mean to suggest that staff members’ donations were being made for the purposes of storytelling; but rather that for most staff members, the major sacrifice is not this year’s donation but rather this year’s work at GiveWell. In other words, regardless of how GiveWell employees spend their cash (unless perhaps if they live in destitution or have poor career prospects otherwise), the majority of their charitable gift is made in the form of discounted labor for GiveWell. Any cash on top is just a surplus.

    To me, the storytelling aspect comes in, at a minimum, because to the extent that staff members choose to give to GiveWell, they could just negotiate lower compensation instead, with favorable tax outcomes.

  • Ian, I don’t really agree with this. Working at GiveWell is the result of a complex, multidimensional decision. People spend a lot of time at their job; the environment, their attitude toward the work, etc. are major factors in addition to pay. I do think all our employees could earn more in the for-profit sector, but that doesn’t mean it’s accurate to think of their decision to work for GiveWell as a “donation.” And asking for pay to be lowered as a “donation” would mean (a) losing privacy re: the size of the donation; (b) having to “give” to GiveWell rather than another charity (asking for the “donation” of lower pay to be regranted is not something we’ve checked out all of the ramifications of); (c) missing an opportunity to “eat one’s own dogfood” by going through a similar psychological process to what the donors we’re serving go through.

  • Anjali Manly on December 22, 2013 at 4:06 pm said:

    This is the first time I have looked at GiveWell, but was intrigued by the concept after reading a few articles in regards to your mission. I am a very analytical person, and appreciate the idea of maximizing the effectiveness of the donated dollar.

    I have to say, I was shocked and disappointed in this post. How can so many of you, as staffers who study giving day in and day out, postpone giving to the next year? And how can GivingWell recommend a charity, but yet, your staffers are not confident enough in the charity to donate their personal money? Clearly, there are problems surrounding your recommendation of SCI if so many staffers do not have confidence in their dollars being effective with that organization. How can you ask us, the general public, to donate to a charity, if you won’t as well?

    Back on the topic of postponing donations; I understand that many organizations are not large enough to handle an inundation of money in a single year. But why is that a reason to postpone donating at all? There are more charities that are in need, beyond the 2013 recommendation list. Where there not organizations that didn’t make the list, but still provide an effective use of your donated dollar? What about charities that were listed on your 2012 list? Perhaps those organizations could benefit more from your giving this year, now that the hoopla has died down from making the earlier year’s list. When you postpone donations, do you keep a running tally of the money that you had allocated, but did not give? Or does it just get lost in the shuffle of next year?

    While I am a detail oriented person, perhaps the staffers at GivingWell are getting too caught up in the details, and might be missing the bigger picture of giving.

  • Alexander on December 23, 2013 at 9:39 pm said:

    Anjali – thanks for the comment. First, I’d note that all staff members decided to give at least a large portion of the total amount that they were planning to give this year to our top charities; I don’t think anyone decided to wait with their entire contributions. Second, while I can’t speak for other staff members, I think the case for waiting is not based on the idea that there are not good opportunities now–we strongly believe there are–but rather that there’s a high chance of an even better opportunity coming along within the next year or so. For me, though perhaps not for other staff members, that’s motivated largely by the hope that the Against Malaria Foundation (last year’s top charity) will have more room for more funding in the next ~6 months. Rather than giving the funds now, though, I’d like to be able to wait and see; if they don’t, l’d likely give the funds either to this year’s top charities or next’s. Finally, at least in my case, I’m very careful to set aside the funds that I plan to donate next year as roll-over from this year in a segregated account, so it’s quite easy to track and make sure that they’re not “lost in the shuffle.”

  • Ian Turner on December 24, 2013 at 11:13 pm said:

    Hi Holden,

    It was not my intenion to suggest that Givewell employees made their career choice based on strict utilitarian grounds; of course there are a large number of factors behind one’s choice of what work to do and where to do it. (I’ve heard one’s relationship with one’s boss is one of the biggest factors in individual happiness, although I have no idea if that’s true or not).

    That being the case, I doubt many at Givewell chose the job for the workplace relationships or short commute (and I also recognize I could be wrong here). A desire to do good may not be the only factor, but it has to be a large one to justify the tradeoffs involved. That’s what I mean when I say that “the major sacrifice is not this year’s donation but rather this year’s work at GiveWell”.

    By the way, I am not actually “asking” for anything in regards to this conversation; I just want to be clear that from my perspective, Givewell’s employees deserve recognition for the everyday work in their job moreso than for their end-of-year donation, however large or well-considered.

    I agree that the process of “eating one’s own dogfood” is a useful mental exercise; but it’s not clear to me why the current process (of taking GiveWell money, subtracting FICA and other taxes on the way to an employee, and then transferring it back to Givewell) is necessary to meet that end. But I have nothing to say on the privacy front; if it’s important to employees that GiveWell not know how much of their income they are giving away, even to Givewell itself, then there is nothing to be done about that. Is that the case? The fact that all staff chose to participate in this blog post suggests otherwise to me, but I recognize I’m bad at understanding others’ motivations.

  • Ian, I appreciate your recognition of our staff, but I frankly disagree that the “sacrifice” framing is appropriate in all cases. I really don’t think of the income I forgo in order to work at GiveWell (rather than, say, a hedge fund) as being comparable to money I choose to give to charity. The former decision feels self-serving to me (I’m glad to forgo the income because of other, less tangible benefits) in a way that the latter does not. This only addresses one employee (myself) and obviously others may feel differently, but my point was to articulate why I don’t think we can safely assume that framing is appropriate for all.

    Using payroll deductions for donations is something we simply haven’t brought up in a serious way. The idea has occurred to us, but in order to make this an option for employees there would have to be a fair amount of discussions and investigation that we just don’t think are worthwhile. Possible downsides include the ones I named and probably some I’m not thinking of. The current setup, in which everyone directly gives if they want and what they want (as people who don’t work at GiveWell do), is easy, requires no special setup, puts our employees in the same position as our “customers,” and is overall good enough that we aren’t interested in pursuing another setup.

  • Ian Turner on December 28, 2013 at 10:56 pm said:

    Holden, thanks for your thoughts and perspective. It sounds like you disagree with Dan Palotta’s analysis here, where he describes earning to give as the more self-serving choice. Is that right?

  • Ian, I’m not making a general comment about whether working at GiveWell is best thought of as “selfish” or “sacrificing”; my point is to say that no generalization is safe, and the situation is different for each individual.

    For me personally, it doesn’t feel like a sacrifice. I value the things I’ve gained more than the things I’ve given up.

  • Karen Jeffrey on January 26, 2014 at 5:15 pm said:

    I think this post is a great idea in principle, but I’m afraid that the content of it has done more harm than good in terms of my confidence in GiveWell. I don’t say this in order to gripe, but rather so that you can be aware of this negative impact, and avoid it if possible. I think that rigorous and transparent evaluation of charities is highly valuable, and I make this comment in order to help GiveWell to preserve or add to that value.

    I second the concerns raised by Anjali Maly in the comments. I find the fact that more than half of the staff have decided to postpone their giving disconcerting. Doesn’t citing the rationale that “I tend to think that future giving opportunities will be better than this year’s” suggest that you will never get around to giving, since this seems likely to always be true? If members of staff have committed to donate their pledged funds at a within a set timeframe and for specific reason, as Alexander has clarified, then stating this explicitly would give me much greater confidence in GiveWell’s integrity.

    The conflict here with the message in GiveWell’s “Give now or give later?” piece, which estimates “that the conceptual ‘interest rate on empowering people’ is higher than the interest rate you earn when you save” also needs to be clearly addressed to avoid undermining your own work.

    I also second Anjali’s point that with the exception of Sean, possibly Timothy and possibly (though it is unclear to me from the post) Jake, the staff are not giving to two of the top three charities recommended by GiveWell. There are several comments in the post which seem to me to suggest a low level of confidence in GiveWell’s own recommendation, including:
    “I think of deworming as an intervention whose main case rests on two problematic studies.”
    “This is a highly non-robust calculation”
    “In the case of deworming, I think there are too many unknowns to put weight on our “best guess””
    “The reasons that I’m not donating to SCI include recent concerns about evidence of impact… [and] seems to have relatively low upside”

    Can it be that of all the charities considered, after narrowing them down to three, there is really only one charity (Give Directly) that the majority of staff are confident enough to endorse with their own giving? If that is the case, perhaps only one charity should be recommended, or a measure of confidence should be included alongside each recommendation.

    I also wonder why the staff aren’t diversifying their giving more, in order to minimise risk in the face of uncertainty?

    I realise that this post is quite critical, but I feel that it’s important to make you aware of the impression that it had on me so that you don’t undermine your own, very important, work.

  • Alexander on January 27, 2014 at 3:24 pm said:

    Karen – thanks for the feedback.

    First, I just want to amplify what I noted to Anjali – all staff members are giving a large portion of the total amount they planned this year.

    Second, I don’t think waiting with a portion of this year’s giving “suggest[s] that you will never get around to giving,” and would note that every staff member who is saving some portion of their giving for the future has explicitly considered this. Some of us who are waiting with parts of our donation (me, Josh, and Ben to some extent) explicitly expect to give the rest within a year, others who are waiting (Howie and Ben to some extent) haven’t specified, and Jake is explicit that he may want to save for more than one year because he’s still relatively new to philanthropy. I think it’s very possible to have reasonable disagreements about these decisions–that’s why we wrote this post–but every staff member thought hard about exactly why they were making the decision they did. Similarly, we’ve all grappled with the argument that the high social “interest rate” to giving means that you should give more now. I think there’s something exceptional this year (I think there’s a good chance of significantly better giving opportunities in a year, which I wouldn’t have said last year); Jake thinks he’s on a steep part of a learning curve. For us, these considerations overwhelm the one you mention, but I certainly agree that it’s worth taking into account.

    Third, we don’t think of our own giving primarily as endorsement. If we did, I believe all staff would give to all the top charities. Instead, I believe we think as hard as we can about where our money will do the most good, and give there. For most staff members this year, that meant donating to GiveDirectly, but for some that meant splitting donation with other top charities. This decision was predicated on the view that absent our own donations, we expected all of the organizations to hit the minimum funding targets we wanted them to get, so the relevant question for the individual was where to direct marginal funding. (On the SCI evidence in particular, there are real concerns we have and we don’t want to hide them, but we think the overall case for SCI is a lot stronger than that of most charities, just (for some of us) not as strong as the case for GiveDirectly.) On diversifying to reduce risk in charity, see our discussion in the “How much should you give to each charity?” section of our 2012 top charities post.

    Stepping back, I want to reiterate that I appreciate the feedback on this post. Your response and several others we received suggest that this post didn’t have the clarity that we normally aim for on the blog. GiveWell staff take the questions of where, when, and how much to give very seriously: in addition to spending the bulk of our work hours trying to answer the “where?” question, we all spent a long time thinking about “when?” and “how much?” As a result, I think we did a worse job than we should have situating our reasoning for the general blog audience, and I apologize for that.

  • Karen Jeffrey on February 4, 2014 at 5:44 pm said:

    Thank you for your detailed response, and I?m sorry that?s it?s taken me so long to get back to you.
    Ok ? great, as long as there is something in there to clarify that the giving won?t be postponed indefinitely (because there might always be a better opportunity in the future), like committing to give within some defined period, or rolling over the giving in a cumulative way, then I think you?ll be setting an excellent example for all of us!
    Thank you for clarifying on the ?social interest rate? point.
    What you?ve said about the SCI evidence is helpful and sounds sensible. I guess for you, who are so immersed in this research, the fact that ?the overall case for SCI is a lot stronger than that of most charities? is obvious, and so it?s easy to forget to qualify the concerns you express about SCI too; but for those of us not immersed in this field, that qualification is important ? otherwise we feel puzzled about a message that seems conflicting (e.g. we have concerns about the research vs it?s one of our top recommended charities).
    I really appreciate you taking these comments on board, and I?m sorry for being (probably a little too) harsh. Your response has reassured me that my complaints were over nothing more than a communication issue.

  • Peter Hurford on February 5, 2014 at 9:26 am said:

    Karen, I don’t think GiveWell employees are not giving because of the mere *possibility* that *sometime* in the future giving opportunities *might* be better, as you imply. Rather, it’s an explicit calculation that the impact of donating more in the future exceeds the impact of donating now plus the compounding interest from donating now plus the social benefits from donating now. It’s unlikely that such will be the case very often, but given current circumstances with AMF, it seems quite plausible.

Comments are closed.