Elie recently highlighted his doubts about our top charities, and a commenter responded:

Of course one ultimately never knows how much good a charity or any given donation will do. Bednets might end up saving the life of child who grows up to be the next Nelson Mandela – or the next Saddam Hussein. Everything we do is a gamble, but I’d like to make the best ones I can. These two charities look like good gambles.

I agree completely, but I’m still glad Elie emphasized these risks. Because in one specific way, supporting our top charities is riskier than supporting any other charity: if something goes wrong with one of our top charities, it will come out promptly and publicly.

There’s no question that “giving a goat” has caught on, as a way of getting people to support charity – rather than consumer goods – for the holidays.

I certainly don’t need an iPad or a new TV, and I encourage friends and family to make donations in my name rather than sending me gifts. However, I’d rather not get a “goat” because:

• I’d have unanswered questions about the effects this gift might have. Will the person who receives the goat be well-suited to receive it? Will they mistreat the goat? Will the goat produce enough to justify their investment in it? Will the local community perceive unfairness in who is chosen to receive the goat?We’ve found livestock-gift programs to be among the more poorly documented developing-world aid programs out there. We have little sense of whether and when these concerns apply.
• The model of giving out goats fundamentally doesn’t make much sense to me. We wrote in 2009 that giving out livestock seems to have the same challenges as giving out cash, plus additional challenges. This isn’t to say that it’s never a good idea. But supporting a model with such fundamental questions would make me uneasy unless I saw those questions being addressed in an intelligent, context-specific way. (Note that there is a charity giving out cash.)
• It’s not really a goat anyway. See our 2009 discussion of donor illusions: chances are, the fine print says you’re really just giving a donation to the charity, not any specific goat. Matching your dollars with a specific goat doesn’t (in our view) really make sense logistically, so we’re glad that it isn’t actually happening.
• I think there are much better ways to help people with donations. This is the most important point for me. Maybe “giving a goat” really does help people – but I don’t believe it helps them as much as possible given our other options. GiveWell’s top charities carry out proven, cost-effective interventions; they have strong track records; they have better transparency and accountability than any livestock-focused programs I know of. They can generate incredible value for your money, such as saving a life for $2000 or keeping someone worm-free throughout childhood for $5. Those are our best evidence-based estimates (the details are all online), not marketing pitches.
• Giving a goat means supporting the best story for donors, not the best solution for the people I want to help. If you agree that there are better ways to help people than “give a goat,” you should be worried about the message you send when you opt for the latter. To me, the whole point of “giving to charity for the holidays” is to take the focus off of our own wants and put it on helping people who have much more fundamental needs. This goal is cheapened if we support any charities other than the best we can find.

We like the idea of giving to charity in someone’s name as a gift for the holidays. We’ve recently made this possible for our top charities. We just think these gifts should be made to the best charities possible.

My favorite cause used to be U.S. equality of opportunity. But over the years since we started GiveWell, I’ve become more and more convinced that the best giving opportunities for individual donors lie in the area of global health and nutrition.

Fundamentally, this area stands out because it involves a lot of interventions that have measurable, demonstrable, quantifiable benefits, yet also haven’t been funded to reach everyone who can benefit from them. (See our list.) It’s the only aid area I know of fitting this description. This has made charities working in this area a good fit for our criteria.

One might retort that this just shows our criteria to have the problem of focusing on what can be measured, rather than on what’s best, but I would disagree. I think that there are good reasons to believe that the cause of global health and nutrition really does contain the best – and not just the most easily understood – giving opportunities.

• Global health and nutrition interventions have the most impressive track record in international aid, with the possible exception of funding scientific research. I believe (though we are still investigating this) that the same holds for aid in general. To some extent this may be an artifact of how measurable the goals are, but it’s still a point in this cause’s favor.
• Global health and nutrition interventions are incredibly cheap on a per-person basis. We don’t have “cost per life saved” type figures for any programs outside of health, largely because the effects of these programs aren’t well-known enough. But consider that
  • An insecticide-treated net costs under $6, and a deworming treatment costs around 50c – these figures include all shipping costs, administration costs, etc. By contrast, GiveDirectly’s cash transfers cost $500 per household per year; the promising Ultra Poor Graduation program appears to have similar costs (PDF, pg 3); educating a child, even in the poorest countries, costs $50+ (sometimes much more) per year (PDF pg 39); even in microfinance, $12-17 per year is spent in grants alone for each current person who is served by microfinance. Programs that don’t target the poorest of the poor – such as U.S. programs – cost far more. These are far from apples-to-apples comparisons, but my intuition is that the health interventions compare favorably on “quality-of-life impact for the money.”
  • Speaking more informally, on our site visits it has struck me how lean health interventions often are. Because the evidence is so strong, people generally know exactly what has to be delivered, and how, in order to benefit people. This means they can design programs to deliver these crucial items and services with maximal efficiency. Other programs are much more intensive – again, for not clearly more life benefit. This observation makes the above point make more sense to me: health interventions are cheap because they are small and precisely defined but involve major known benefits.
• Health and nutrition programs are particularly good at having clear goals and accountability, leading to learning over time. Because benefits can be quantified and predicted, targets can be set, and adjustments can be made when they’re not met. We’ve been placing more emphasis on the criterion of accountability, i.e., the likelihood that giving will lead to learning. Health and nutrition interventions stand out in terms of the likelihood that giving will lead to learning.
• Room for more funding is relatively easy to gauge in the area of health and nutrition. We believe that one of the thorniest issues that a donor has to contend with is that of room for more funding. In an area like funding scientific research (which otherwise has much to recommend it as a charitable cause), it’s particularly hard to answer the question, “How does the value of the next study on the priority list – the one that no one else has funded and that my funding will make possible – compare to the value of the average study overall?” This isn’t nearly as much of an issue when delivering global health and nutrition interventions; the expected benefits of running an additional project tend to be easier to quantify.
• Health and nutrition are fundamental to quality of life. In my view, the value of additional education – or additional income – depends a lot on the context, but health and nutrition are fundamental and universal needs. In addition, with economic empowerment or education interventions, I tend to worry about the difficulty of distinguishing “zero-sum benefits” (helping some people at the expense of others) from “positive-sum benefits” – a well-intentioned business training or education program could end up simply transferring power/wealth from some people in a community to others, and this could be very difficult to distinguish from actually improving these people’s productivity. I feel this is much less of a concern when it comes to health and nutrition.

What problems are best suited to having money thrown at them?

I’m not arguing that health and nutrition are the most important issues in the world. There are many changes in the world that I’d like to see and that I believe are worth fighting for. There are many fantastic uses of money outside the sector of global health and nutrition; many involve taking risks on building institutions and producing public goods. But when thinking about how individual donors can do good, my mind jumps to the easiest ways to have impact without having special insight – the problems that are most likely to be solved by “throwing money at them.” When this is the goal, being able to measure, demonstrate and quantify one’s impact is enormously helpful.

According to our analysis, ~$5 can buy a bednet or 10 years of deworming – either of which will have a substantial, quantifiable, definite impact on quality of life. That can empower a person who is better positioned than you are to address many of their other problems. I believe that for a donor interested in making the world better just by writing a check, this sort of value-for-money may not be available in any other cause.

People sometimes ask us whether they should give now, or save their money and give (including the interest/returns they accrue on their money) later. We don’t think there’s a clear answer. Here are the major issues as I see them, when thinking about my own giving. Bottom line – my favored strategy at the moment is to give regularly (a set percentage of my income each year).

Will tomorrow’s giving opportunities be better than today’s?GiveWell’s progress may lead to better giving opportunities in the near future. I believe GiveWell’s research is the best available for identifying great giving opportunities; but GiveWell is still a very young organization, and we still have a lot to learn. It’s very possible that we’ll find much better giving opportunites in the future.

On the flip side, our growth and learning depends on growing our money moved. Giving to our top charities today helps this happen.

None of the best giving opportunities I see today are guaranteed to be good opportunities next year. In some cases (particularly VillageReach in 2010), giving now may be crucial to an organization’s development. Even when this isn’t the case, there’s always the possibility that the organization will improve its fundraising – or make a contact with a large funder – in the future, affecting its room for more funding.

Economic growth, increased giving, and smarter giving may mean that giving opportunities are worse in the far future. We’ve written before about the idea that the more good is being done by others, the fewer opportunities you have to do a lot of good with your own giving. We hope to see a day where no one is so poor/underserved that you can save their life by giving few thousand dollars.

Do I “earn interest” by saving now and giving later?
If I save, I earn interest/returns on my money; but if I give, the good I accomplish may “earn interest” too. Ideally, donations give people more power over their lives, which in turn leaves them better positioned to help others (in their community or elsewhere). We take seriously the idea that most problems may be better addressed by local communities than by outside aid; we focus our outside aid on the problems we feel it is best-suited to address.

In general, I would guess that the conceptual “interest rate on empowering people” is higher than the interest rate you earn when you save. This is because the same basic mechanism underlies both (having more resources today can allow a person to generate more resources for the future), but savers get paid directly for providing resources, while donors don’t (thus, there is likely more “unmet need” for donations than for savings). Given today’s interest rates, it seems particularly likely that the “interest rate on empowering people” is higher than the interest rate you earn when you save.

Other issuesGiving now means “putting my money where my mouth is.” I believe that having my own money at stake helps me think harder, more concretely, and more realistically about where I should be giving and what results I’m expecting from my giving. This in turn helps me learn over time.

I think the same applies, to a lesser extent, to donors who are relying on GiveWell’s research rather than doing their own – giving now may help you think through what questions to check GiveWell’s answers to, and consider how you feel about relying on GiveWell’s research vs. going in another direction.

If I postponed substantial giving for a long time, I’d worry that I was giving myself an excuse not to give at all.

Giving regularly and predictably is more helpful to the group(s) I’m supporting. In general, people tend to give the same donations each year,* and I believe that charities often plan around this fact. (We certainly do, both when considering our operating expenses and when considering our money moved.) Therefore, if you give irregularly (particularly if your donations go down over time), you are sending an inaccurate message to a charity and may be negatively affecting its ability to plan, unless you take specific efforts to communicate your plans (and even when you do so, this may create extra hassle for the charity).

Bottom line. In general, my preferred approach is to give relatively regularly; I think there are substantial disadvantages (discussed above) to deviating from this approach in either direction (giving a lot now and only a little later, or giving a little now with the intent of giving a lot later).

At some point I may spot a giving opportunity that seems like a true outlier, and focus my giving on that opportunity rather than giving regularly over time. To give one example of this – when we started GiveWell, we believed that our startup funds were critical, and thus we encouraged donors to give a large amount immediately even if it meant abstaining from giving for the next year or two. Several did exactly this, and I think in hindsight it was the right decision (their 2007 gifts were in fact critical for our getting off the ground, in a way that they wouldn’t have been a year later). So I do think there can be good reason to “over-give” now and hold back later, or to save now and give more later. But right now I don’t see any compelling argument for deviating from “give a set percentage of my income each year.”

*Our experience suggests this; the Money for Good study suggests it as well (see page 16).