This is a guest post from Ian Turner about how he decided what charity to support for his most recent donation. Ian and GiveWell staff had several in-depth conversations as he worked through his decision, so we invited him to share his thought process here. Note that GiveWell has made minor editing suggestions for this post (though Ian determined the final content).

Every person who gives to charity does so for a different reason, and these reasons are as numerous and as varied as the thousands of causes and millions of charities that pursue them. I’d like to share a little about recent choices I’ve made, and how that approach to giving has developed over time.

For me, the essence of giving comes down to three distinct beliefs or realizations. These are, in turn, the limitations of money; the importance of money; and an actual desire to help others. When you combine these three beliefs with actual resources to give, one inevitably must give and must give for impact – that is, one must give well. Thus my approach. I’m not attached to any particular sector, cause, approach, country, or region. But I do look to maximize the effect of my gifts on the human condition overall.

First gifts: A rough start

In the very beginning, as a child, my father greatly influenced my thoughts about giving. He placed ideas in my head such as “don’t give to panhandlers directly, give to organizations”, and created the legitimacy, if not the expectation, of giving freely. Out of this idea, the desire to help now (and not in the unpredictable future), and out of the tradition of tithing shared by many religions, I decided even before I ever had any real earnings that I would give 10% of income (after taxes) to charity, on an ongoing basis. That is a practice that I have continued in the over a decade since.

I had no idea at the time how hard aid actually is, how unsubstantiated most of the nonprofit sector is, or how hard it is to find an effective organization to give to. My first gift was in September 1999, in the amount of $650 (10% of the income from my first summer job) to an organization picked more or less at random: The Unitarian Universalist Service Committee (UUSC). I didn’t have a clear objective other than improving the human condition generally, for which their mission seemed broad enough, and the only real thing that set them apart is that they made the right ask at the right time.

I was disappointed to receive, over the following year, a variety of communications from the UUSC, indicating that my money was being spent in all kinds of ways that (to me) didn’t make sense, from US political activism to questionable (to me) small-scale village business projects. I resolved that I would do a little better next time.

So the next year, after another (more profitable) summer job, I made a different gift. At the time I figured world hunger was a bigger problem than fair trade for cocoa farmers, but thanks to seeds planted by my father I figured there was a better way to tackle hunger than just handing out food. And that’s how I found Freedom from Hunger. Freedom from Hunger operates a large variety of international programs, the centerpiece of which claims to combine microcredit and education for poor rural women.

To evaluate the organization, my main source of information was annual reports. I went back a few years, but didn’t look at any studies or third-party information. Even with this little information or experience, however, I could tell that there were some differences between organizations and looked at a few different ones. Freedom from Hunger had what seemed to be the most impressive annual reports, which I think means intelligent copy that told a story which seemed reasonable, at least from my naïve perspective. So in 2000 I gave them $2,500.

As an exercise, I recently revisited Freedom from Hunger using my now more experienced perspective, to see what evidence of impact they really had. The answer could be described as more than most, and less than the best. Freedom from Hunger did perform a number of controlled studies examining different aspects of its flagship programs, but the results seem questionable, prone to cherry-picking, and it’s not clear to what degree the evaluated treatments match the programs actually implemented worldwide. Freedom from Hunger’s programs today are also significantly more complex than they were in 2000.

One of the biggest lessons I learned from the 2000 gift is that giving well is hard, harder than I ever would have guessed two years previous. I decided I could manage the process better if I made larger gifts less frequently. Eventually, I settled on keeping 10% of my income in a special account each year and then only actually giving it away triennially. As it turns out, I didn’t have any significant income again until 2004, and for tax reasons my next gift was not until 2006.

As an aside, I should note that the triennial 10% gift is specifically for philanthropic efforts. I also make gifts to my church, to people I know personally who are in need, and to political organizations whose causes I support. But I don’t consider that philanthropy, I don’t make the analysis that I would for philanthropic gifts (which are much larger), and I don’t count such gifts toward my 10% donation.

Seeking a better way

After 2000 I did not give again until 2006, but I was not idle during the intervening years. Through The Economist I learned about two organizations of significance, Pratham (Economist article) and Geneva Global (Economist article). Pratham is an Indian educational charity notable for its built-in evaluation approach and closed-loop feedback process. Although Givewell did not exist at the time I discovered Pratham, it has since been reviewed and recommended by Givewell. Geneva Global is a for-profit company that provides advice to donors, especially in terms of impact prediction and evaluation. For their services they charge a fee equal to 15% of the amount donated. Of course, when the time came I also had another look at Freedom from Hunger. I didn’t have any other way to cut through the million charity options, so those became by default my three options to evaluate.

Of the three, I eliminated Freedom from Hunger early on. By now I saw that there was more to impact than glossy annual reports, and I had a growing skepticism about microfinance generally. I was still looking to address root causes; education seemed like a better bet; and Pratham seemed to be making a big difference there. Geneva Global’s main advantage was an appeal to expertise: Their job was to know where to give, so the only question in my mind was whether their expertise was worth the 15% price tag.

In the end, the choice came down to my own procrastination: I didn’t start contacting organizations to ask questions until December, and by the time I found out who at Pratham could answer my questions, they were already away for the holidays. I couldn’t get any substantive information from Pratham until after the new year – not even basic financial information, let alone detailed information about programs. Geneva Global, on the other hand, was beyond responsive: They had lots of detailed information on specific projects, and best of all, a brand-new fund which was offering a 1:1 match to attract starting capital. So instead of paying 15% overhead for Geneva Global, I would end up getting a 70% uplift. In the end, on December 29, 2010, I put a check in the mail for $14,000 toward Geneva Global’s Rwanda Human Empowerment Fund, a collection of projects focused on education, health, and land rights in Rwanda.

As it turns out, I probably chose poorly in 2006. Geneva Global went through a few different purchases and restructurings before winding down the Fund and sending out a final report. This indicated that the Fund had only attracted about a third of the expected amount, but that Geneva Global had begun spending in the first year on the assumption that additional donations would continue to come in. As a result most projects were cut off or could not be continued in the second and third years of the fund’s operation.

In addition, some projects looked highly questionable to my untrained eye. For example, one school project started construction on new classroms, but then shifted midway to make the building a large $64,000 dining hall instead. The school already had a dining hall; the only benefit mentioned in the report was that all 1,000 students could dine at once rather than in shifts. Another project complained that a grantee “lacks professionalism, transparency, and honesty”, leaving the project unfinished and the funds missing. Roughly 1/3 of the total fund was invested in a land rights initative, but the report diplomatically notes that the grantee, the International Justice Mission, “faced many challenges as an outsider”. Finally, none of the projects seemed to have much in the way of quantitative evaluation; with respect to the health initative, the executive summary notes that “Though no formal study has been conducted on the impact of these improved services on infant and maternal mortality rates, we have no doubt that they consequently declined”.

My overall impression was that Geneva Global’s in-county contractor had been given some combination of too much leeway and too little training, and that a substantial amount of the money was lost to subcontractors, spent on unplanned expenses, or wasted on poorly planned projects.

The main lesson I took away from this is that even with a fair amount of effort, giving for impact is really hard! So for my next gift, made in 2009, I determined to do a better job. I started the process early in the year with only Pratham in hand, but quickly discovered resources like GiveWell and Good Intentions are Not Enough. It seems that over the course of the 2000s, a lot of people got interested in impact philanthropy all at once.

Information: A trickle becomes a torrent

GiveWell turned out to be a critical source of information for my 2009 donation. Besides pointing me to exceptional organizations like Population Services International (PSI; Givewell review) and VillageReach (Givewell review), the GiveWell blog and its archives provided a steady supply of new perspectives on charity and giving, ideas which I had never encountered before. Sure, I had long suspected that there were large differences between charities, and had discovered through my own mistakes that giving was hard, but I had never thought about things like offsetting impacts, the complexities of microfinance interest and repayment rates, the costs of receiving aid, or room for more funding. GiveWell suggested some great organizations and then gave some useful benchmarks by which to measure them.

In the end, I settled on a shortlist of Pratham, PSI, and VillageReach. The latter two I discovered through GiveWell. PSI is a large international health charity with some innovative ideas about how to leverage markets to improve distribution and use of health products and services like condoms, bed nets, circumcision, and vaccination. VillageReach is a US charity originally operating exclusively in Mozambique which is notable for having one of the most complete and inarguable demonstrations of impact among charities of any size.

GiveWell also recommended a few other organizations that didn’t pique my interest as much. The Against Malaria Foundation (AMF; Givewell review) interested me less than PSI primarily because of the latter’s inclusion of reproductive health as an essential part of health care delivery. And the Stop TB Partnership (Givewell Review) is an organization which seems to be very effective, but which I heard about too late to consider for a 2006 donation. In addition, giving to the Stop TB partnership would be significantly more difficult, since it is not a US organization and since my 2009 donation was mostly in the form of appreciated stock.

I contacted all three organizations (Pratham, PSI, and VillageReach) and engaged each of them with a battery of questions regarding their programs, evaluations, use of marginal funds, and much more. Owing to the earlier date (and perhaps larger donation size), I had a very different experience compared to 2006: All three organizations were very much forthcoming with both information and time. I read a lot of reports, came back with more questions, and spent hours on the phone, both with charities and with folks from GiveWell. This was a tough decision, because the organizations are fairly different from each other but each with a compelling case.

Pratham was the first to be disqualified. Essentially, the majority of Pratham’s spending goes to two types of programs: Field programs, where Pratham operates remedial education sessions directly on a village-by-village basis, and Read India, where Pratham partners with local governments to improve the quantity and quality of education in public schools. In an attempt to reach the massive scale required for nationwide impact in India, Pratham has focused most of its growth on Read India. But the best case and evidence, both in terms of program design and in terms of third-party evaluation, is for the field programs. In late 2009, Pratham was in the midst of conducting rigorous controlled trials of a few Read India implementations. I have no doubt that they’ll make a strong showing for my 2012 gift, but to me there was not enough evidence on Read India to justify a gift in 2009.

That left PSI and VillageReach: Both health charities, though with very different scale and approaches, making for a difficult comparison. My decision was made more difficult by the fact that VillageReach was in the middle of its own strategic planning process while my evaluation was underway. Indeed, I had originally decided against VillageReach, but gave it new consideration according to new information provided at the last minute.

PSI is a large organization and a relatively known quantity. They provide a fairly advanced quantitative model attempting to translate any and all health interventions into Disability-Adjusted Life Years (DALYs), then allocating funds to the most cost-effective projects, as measured in cost per DALY. Of course, a large organization like PSI receives a lot of restricted funding, so the organization operates some projects even if they have very high cost per DALY. Nonetheless, PSI prepares detailed and easy-to-read reports on both impact and cost-effectiveness. The questions are likewise fairly straightforward: Is the model which translates pregnancies avoided into DALYs a correct one? Are the projects sustainable? Is PSI simply replacing health products and services that would be provided by free markets or governments, or is it expanding markets to consumers that would not be reached otherwise? PSI’s programs are clear but determining true impacts must unfortunately leave some of these questions unanswered.

VillageReach

VillageReach is another story. In order to understand the organization, you need to understand its history. VillageReach aims to improve health outcomes by making distribution systems for health products, especially vaccines, work more efficiently and effectively. The organization started its programs in 2002 with a pilot project in the Cabo Delgado province of Mozambique. VillageReach created a new approach to vaccine distribution, shifting the burden from health clinic workers to a set of staff and equipment dedicated to distribution. In addition to delivery personnel, VillageReach created a new job position to monitor the inventory of vaccines and the effectiveness of the delivery schemes. All these people nominally worked for the government, and all this equipment nominally belonged to the government, but VillageReach paid all the costs, whereas costs for pre-existing systems had been covered by the government. In order to make all this work, VillageReach had to create a whole new distribution company for propane gas, which was previously prohibitively expensive to purchase in Cabo Delgado. VillageReach reports that the company, named VidaGas, is now self-sustaining.

The pilot program ended in 2007, at which point VillageReach handed control over to the government, which promptly dismantled VillageReach’s work and returned to the previously extant approach. But in the meantime, VillageReach prepared two significant reports: One on impact and the other on cost. The two reports compared the cost and effectiveness of the VillageReach approach to the one the government had been using previously, by comparing vaccination rates in the trial province of Cabo Delgado to those in the neighboring province of Niassa. The conclusion is as powerful as it is inescapable: When compared to its neigbor, Cabo Delgado saw much higher rates of vaccination at significantly lower total cost. In other words, the VillageReach approach was lower-cost, not only on a per-vaccination basis but on an overall basis. Indeed, the vaccination program was so cheap that as a donor it would be considered a bargain even considering that VillageReach had to displace all the government spending.

This is the history I had in hand when I started talking to VillageReach in 2009. And when I asked the question, “what’s next?”, I was concerned about the answer. Of course, the vaccination program in Cabo Delgado had already been shut down for two years, and in its place VillageReach talked about a variety of consulting and advocacy programs. But I had no evidence about VillageReach’s effectiveness as council in any country, let alone in those far away from Mozambique. Essentially, given the future plan VillageReach had presented to me, I had to reject them: All the evidence of effectiveness was in relation to a program that was no longer operating.

So I decided to give to PSI. But I didn’t want to give them a free pass, so I resolved to evaluate them as thoroughly as I could, quickly compare that to some other organizations’ claims, and if PSI seemed adequate to choose them.

But, at the last minute, due in part to urging by GiveWell, I gave VillageReach one last phone call, wherein they explained a new agreement with the government of Mozambique to roll out the Cabo Delgado vaccination distribution approach nationwide. This seemed to me to be much closer to VillageReach’s proven pilot project and thus much likelier to be a success. The pilot project had been so successful that I couldn’t say no to a nationwide deployment, and so I gave my $35,000 2009 contribution to VillageReach. I later found out, much to my surprise, than this gift was and is their largest ever donation from an individual.

Back to the Future

Which brings us to the end of the story (for now). In an adventure spanning a bit over a decade, I’ve learned a lot of surprising things: Doing good is harder than it looks. Good intentions are not enough. Helping others is rewarding. And maybe, with the careful application of analytic thought, giving money really can make a difference. Like science, it’s a process of discovery. You try something out, you look to see how it goes, you learn something, and then you repeat the process, and hopefully your little extra knowledge and experience lets you see something deeper the next time around. It’s not easy, but it is possible.

As I look forward to the future of philanthropy, both mine and others’, it’s an exciting picture. There will be new discoveries about the nature of giving, and as the nonprofit sector grows daily more interested in measuring impact, the number of opportunities to give effectively grows with it. And I’m confident that my own skills at giving will continue to grow as well. Charity can change the world: Just make sure you look before you give.

Acknowledgements

This essay would not have been possible without the thoughts, ideas, inspiration, proofreading, and support of a surprisingly large group of friends and family and of the GiveWell staff. You have provided me with essential feedback for which I am unremittingly grateful.

Thanks also to the many people and organizations that have provided me with time, information and ideas throughout my philanthropic career. A donor can accomplish nothing without a place to give, and I am grateful for each of you, even when I have directed funds elsewhere. Our field is in an interesting time at the moment, and is well-poised for significant new accomplishments. Let’s make it happen.