I had a typical reaction to the disaster in Myanmar: wanting to do something. I have spent very little time looking into the area of disaster relief, so after a bit of Googling and discussion with Elie, I gave to Population Services International for two reasons:

• PSI was the winner of our “saving lives” cause for 2007; we are extremely impressed with the organization as a whole, particularly its commitment to thorough self-monitoring. We don’t know much about their relief operations, but I would bet on PSI over any other international relief organization I know of just in terms of the extent to which it “runs a tight ship” with solid monitoring and oversight that allows accountability from the field to the top.
• PSI has a major and long-established presence in Myanmar; I believe (based mostly on this article) that having a pre-existing presence is important, particularly in a situation like this where the idiosyncracies of the area and particularly government seem important. I’m most comfortable with an organization that is used to getting work done in this political and cultural environment.

This is an informal, personal recommendation; it is backed not by an in-depth research project, but by the quick heuristics above.

This also got me thinking, though, about the more general cause of “disaster relief.” We looked into this cause back in 2006 (when we were still a part-time group of volunteers) and found very little. We aren’t aware of any organizations that are exclusively committed to disaster relief; rather, it seems to us that most relief efforts come from large humanitarian organizations, such as PSI, the Red Cross, World Vision, CARE, Direct Relief International, etc. that spend most of their time and money on direct, day-to-day (not disaster-related) aid. This makes sense, since it means emergency aid efforts can be aided by already-on-the-ground presences.

However, it isn’t necessarily the case that the best “day-to-day” relief organization is the best disaster relief organization. The former may be best accomplished through meticulously planned long-term projects that rely on proven techniques to get the maximal dollar-for-dollar impact; by contrast, I would guess that a disaster presents problems that are unusually simple to solve (people who need basic supplies, but who don’t necessarily suffer from a host of interrelated physical, economic, and cultural obstacles), and that speed and efficiency are more important. I’d be very interested in a compiled summary of disaster relief efforts over the last 10 or so years – which organizations were first, and most instrumental, in each relief effort. It seems feasible that such a summary could be created by polling affected governments and citizens, but I’ve never seen one.

I also wonder whether there are cost-effective “disaster preparedness” measures that can aid particularly vulnerable areas in advance. I was shocked at the death toll from this particular disaster, and I wonder whether a similar storm in the U.S. could have been nearly as devastating. It’s possible that disaster preparedness comes mostly from widespread economic prosperity, and that nonprofits are ill-equipped to bring about the kinds of drastic changes that would be needed to improve preparedness (and/or that the areas least equipped for disasters also have other, more important problems). But it also seems possible to me that constructing some extra shelters – or equipping communication infrastructure to provide effective early warnings – could save lives far more effectively than focusing only on after-the-fact interventions.

Looking into these questions, as with just about any area of philanthropy I can think of, would take significant time and resources. I’m not sure whether we’ll get to do it anytime in the near future. But it seems likely to me that the costs of such investigation would be more than justified. When disaster strikes, a lot of people reach straight for their wallets, and give without having time to think about their different options. But the thinking could be done, centrally, in advance – imagine what a difference that would make.

We’ve wondered why scholarship programs don’t have a stronger impact on academic achievement, and have guessed that it’s because disadvantaged children are so far behind by age 5 that they need special schools, with a special approach, if they’re to have any hope of catching up.

The quote below, from an article in the Washington Monthly (h/t Kevin Carey), offers another possibility: the private schools students with vouchers attend may be little better than the public schools they leave. This is a report on the Milwaukee voucher project, not the New York programs we’ve focused on, but it makes me wonder if New York private schools could be as troubled.

In 2005, a team of reporters from the Milwaukee Journal Sentinel visited all but a handful of the private choice schools, and found that “the voucher schools feel, and look, surprisingly like schools in the Milwaukee Public Schools district. Both … are struggling in the same battle to educate low-income, minority students.” The Journal Sentinel also reported that the absence of oversight from the much-derided government bureaucracy had led to a significant waste of public funds, and even outright fraud. At least ten of the 125 private schools in the voucher program “appeared to lack the ability, resources, knowledge, or will to offer children even a mediocre education.” Most of those schools were led by individuals who had negligible experience and had no resources other than state payments.

In response to Elie’s recent series of posts on malnutrition, John J. comments:

“What specific nutrients are they missing…etc etc etc?”

Here’s a question for you: What possible use is this question to Givewell? Do you really need to get into such miniscule depth with regard to poor people who can’t afford enough food and who are malnourished as a result?

One might accuse you of wasting the time of non profit workers with such picky detail. Serously, I’m not just being cranky here–what was the point of these questions? It seems to me that common sense is enough here: people don’t get enough of the rights kinds of foods to eat, and helping them get enough is…helpful. Really, what was your point? And, additionally, what was the point of this post at all?

First of all, we don’t believe that food aid is necessarily helpful: we’ve seen plausible arguments that it can do more damage (by undermining local farmers’ business) than good. (See this critique from Philanthropy Action, co-maintained by Board member Tim Ogden, as well as this story on CARE’s decision to withdraw from the US government’s food aid program.) Broad enough outcomes data could mitigate this concern, as could clear information on the local food market in the region in question; without it, we’d still bet that food aid is a good thing on average, but could easily be wrong.

But the reason we ask such specific questions isn’t primarily to determine whether aid helps; it’s to find the aid that will help as much as possible (in ways that fit our philosophical goals). Like any donor, we choose between literally thousands of charities; a core idea of GiveWell is that under these circumstances, it doesn’t make sense to settle for “some” good accomplished.

Given the variety of different approaches to malnutrition, we expect that different charities vary wildly in both:

• What kind of life change they’re bringing about. For example, Vitamin A deficiency may significantly increase the likelihood of death before age 5, while deficiencies in Iron and Iodine may lead to Anemia and reduced I.Q. These are fundamentally different benefits that can’t be reduced to the same terms – and that different donors will value differently. In order to understand our options, we need the details of what sorts of malnutrition are being addressed, and where.
• How many lives they’re changing (i.e., cost-effectiveness). We find it possible that some charities are simply carrying better-conceived and -executed programs than others – that means more people helped, for the same funds. And even if different malnutrition programs turned out to be roughly comparable to each other, we’d still want to know how they compare to all the other health interventions out there, from hospitals and health centers to condoms and bednets.

These issues don’t matter very much if the only line you draw is between “donation was squandered” and “donation helped people.” But if you want to help people as much as you can, the lack of public answers to our questions is a real problem.

Below is the email I received from Kelli M. Donley, Program Director at Esperanca, in response to the questions I asked in Part I and Part II of this series:

Dear Elie,

Hi there! I am going to try to answer your questions to the best of my
knowledge. I hope this helps:

These questions apply to your malnutrition programs in Mozambique and
Nicaragua (except where noted):

– Is the problem “receiving enough food” or receiving the right types of
food (Mozambique)?

Both. There is not enough food in Mozambique at the moment because of
poverty, the way the population has changed with the civil war, etc. Also,
the food they are able to grow is limited. Where we work, it is
predominantly swap. So, they grow corn, rice and millet, but have little
variety to the fruit and vegetables they are consuming.

In Nicaragua, poverty and environmental concerns have limited the diet and
access to a balanced diet.
– What does Esperanca teach community members? Which foods to plant? How to
cultivate and plant them?

In Mozambique we focus solely on three areas: cholera, malaria and HIV
prevention and treatment. Our partnering agency in this country – Care for
Life – has a small farm where they teach small animal husbandry, improved
seed cultivation, improved environmental techniques to prevent erosion, etc.

In Nicaragua, we have seed banks that promote long-term growing of healthier
crops. We also focus on environmental concerns, as in how to prevent soil
erosion. I like to think our midwifery training program also helps minimize
malnutrition by working with pregnant women to teach them to breastfeed and
why it is important.

– How often do they hold classes and how long each does each last?

This varies by country and by community. These classes are run by the
nonprofits we partner with in the country.
– How well do community members retain and implement what they’re taught?

They regularly attend courses and are putting this knowledge to use
immediately in their communities. One would hope they are retaining a lot of
it.
– What nutrients do the foods contain and how does this match up to the
nutrients community members likely lack?

I do not have this information to share with you. I can tell you that meat
and dairy are extreme luxuries in both areas. Fresh fruit and vegetables are
often limited in scope and supply.
– How effectively does this program reduce malnutrition of specific
nutrients?
– What effect does this have on the outcomes (e.g., mortality rate, general
health, I.Q.) of community members?
– How many people do they help?
– How much does this program cost?

Our programs are not broken down into these sorts of statistics for
analysis. We have general overhead costs for each project. We reach 11,000
people in Mozambique and 30,000 in Nicaragua each year. One would hope we
are improving their quality of life and their life-spans, although we do not
employ any sort of researcher to determine this. It is all through
word-of-mouth and monthly reporting from our field officers that we see
results.

I also have some questions about programs to fight malnutrition in general:

Again, I don’t have the information to answer these specific malnutrition
questions. I hope what I’ve provided will be of some use!

– What specific nutrients are people missing that affect them? Without
knowing this, there’s no way to decide (regardless of the type of
intervention listed above) which nutrients are the most important. Is it
vitamin A? Iron? Iodine?
– What happens because of each type of deficiency (e.g., anemia because of
lack of iron, blindness because of lack of vitamin A, death)? How likely are
each of these outcomes given a particular deficiency? Understanding this
will help me decide which sort of program is most appealing to me – not a
fully scientific/objective decision, since my view of the “good life”
affects which disorders I most want to address.
– In what region do the people most affected live? What age are they? Any
special circumstances (i.e., pregnant women)? Before focusing on helping a
group of people, I think it’s important to know what other obstacles they
face. I want to help people for which malnutrition is the main (or a
significant) obstacle to living a full, happy life. If the people face many
more obstacles (e.g., other diseases, war, etc.) helping them here may not
do as much good.

In Part I, I listed some questions about malnutrition that I think you’d have to answer to have any sense of what you’re accomplishing with your donation, and how it compares to your other options. I then went on a search – similar to what we did part-time before we started GiveWell – to see what sorts of answers, and organizations, I could easily find.

Googling “Charity Malnutrition” doesn’t turn up anything that looks like an obvious first choice, and results from a Charity Navigator search for “malnutrition” provide nine organizations, five of which serve people in the United States.

We’ve already looked closely at Helen Keller International (click for more info), and the mission for the Children’s Shelter of Cebu doesn’t sound like what I’m looking for. This leaves Curamericas Global and Esperanca, two organizations that, based on their mission statements available on Charity Navigator have extremely broad mandates to improve health in the developing world. I chose to look into Esperanca.

Esperanca is a relatively small organization, with expenses, according to its 2006 IRS Form 990 (free registration required at Guidestar) in 2006 of $1.8 million.

Esparanca’s website provides information on their ongoing projects in Bolivia, Mozambique, and Nicaragua. The programs in Mozambique and Nicaragua explicitly address malnutrition:

• Mozambique: “In conjunction with Care for life, our public health programs in Mozambique include:Teaching community members how to make sure their families are receiving enough food”
• Nicaragua: “One of our foremost public health concerns in Nicragua is nutrition. We provide the necessary supplies and training to families and community groups to start neighborhood gardens. These gardens grow quickly and supplement traditional diets with much needed fruits and vegetables. Malnutrition accounts for 54% of all deaths of children under age five, (WHO).”

Esparanca’s Annual Report doesn’t add any information about the programs in these two countries.

The only program related information in Espranca’s 2006 IRS Form 990 is in response to the IRS requirement (in Part III) that “all organizations must describe their exempt purpose achievements in a clear and concise manner.” Esparanca responds, “Provide medical services and public health programs in the poorest areas of the world throughout North America, South America, and Africa, serving 100,000 people.” (Box a)

Given that information, I’m left with a set of unanswered questions. I emailed these questions to Esperanca (they responded yesterday, though we aren’t going to post their response until/unless we get their permission). The questions that follow apply to both programs, except where noted:

• Is the problem “receiving enough food” or receiving the right types of food (Mozambique)?
• What does Esperanca teach community members? Which foods to plant? How to cultivate and plant them?
• How often do they hold classes and how long each does each last?
• How well do community members retain and implement what they’re taught?
• What nutrients do the foods contain and how does this match up to the nutrients community members likely lack?
• How effectively does this program reduce malnutrition of specific nutrients?
• What effect does this have on the outcomes (e.g., mortality rate, general health, I.Q.) of community members?
• How many people do they help?
• How much does this program cost?

Note that these questions aren’t just about the scientific/factual details of malnutrition; they’re about the organization’s competence, effectiveness, and even philosophical priorities. It seems clear that two equally legitimate and well-meaning organizations could produce wildly different answers to them, and thus could be diverging wildly in terms of how what they do fits with my goal of helping people. That’s why I think it’s important to answer them. And I can’t find answers anywhere on the website, Annual Report, or tax forms.

I also checked out CARE’s nutrition page, but it does not help with the questions above. CARE’s website has a searchable database of ongoing projects, but the example project I looked at only provided information on the country in which the project takes place, the goal, and the broad method for implementation (in this case, flour fortification). That’s all somewhat helpful, but in order to donate to this CARE project, I’d have to do a lot of independent research along the lines of the questions I listed above.

My intent here isn’t to criticize the charities above – the information they’re giving may have more to do with what’s being asked for than with what they have. But regardless, the fact remains – if I want to help people with malnutrition, at this point I seem to be stuck. Without more research, my only option is blind faith.

In this post, I’m going to take a fresh look at malnutrition, a cause that interests me a lot. I have two goals in mind here:

• Think through the issue, and what I’d need to know to decide where to donate. This is particularly relevant since there’s a good chance we’ll be studying malnutrition in the coming year.
• Give you a sense of what an individual donor would have to do to make an informed giving decision (and how quickly finding information becomes tricky). This process is roughly what we went through almost two years ago, ultimately resulting in GiveWell’s creation.

The problem

Malnutrition is a major debilitating factor in the developing world:

• According to this report from UNCIEF, malnutrition plays a role in 50% of childhood mortality.
• The New York Times has reported that iodine deficiency is “the leading preventable cause of mental retardation. Even moderate deficiency, especially in pregnant women and infants lowers intelligence by 10 to 15 I.Q. points.”We’ve learned a bit more about malnutrition (for the charity nerds among you), but … 50% of childhood mortality? Leading preventable cause of mental retardation? 10-15 I.Q. points? Wow. Where should I give?

  Potential solutions

  There are several interventions that seem like they fit with this problem. Each raises its own set of concerns about whether it’s being implemented effectively.

  • Educating people (or helping them) eat a more varied diet. As with all “training” interventions, I’d be concerned about whether an organization is effectively bridging cultural divides and changing behavior, not easy things to do. I’d want to know 1) what’s taught (i.e., does it seem logical that the program offered would lead to the desired effect) and 2) how the organization monitors that those who are taught take action consistent with what’s taught in the class.
  • Food fortified with specific nutrients. This raises some concerns about economic disruption (i.e., competing with local farmers), as well as questions about whether people are eating sufficient quantities to get the nutrients they need.
  • Direct administration of a vitamin supplement. Here the specifics of the vitamin are very important: it’s essential to know the number of doses necessary, how the organization monitors the number of doses given, and the consistency with which the supplement works. It’s easy to imagine a vitamin supplementation program which is ineffective because of wasted supplements (i.e., those that the intended users never take) or too few (or too many) doses given to the same person.
  • Removal of other causes of malnutrition (i.e., treating intestinal parasites). It’s important to know how effective treatments are in the short and long term.

  When trying to combat malnutrition, my aim isn’t merely to help some people imrpove their nutritional status to some extent it’s to help as many people improve their nutritional status to the greatest degree possible. In order to find the most effective intervention, I’d try to answer these questions:

  • What specific nutrients are people missing that affect them? Without knowing this, there’s no way to decide (regardless of the type of intervention listed above) which nutrients are the most important. Is it vitamin A? Iron? Iodine?
  • What happens because of each type of deficiency (e.g., anemia because of lack of iron, blindness because of lack of vitamin A, death)? How likely are each of these outcomes given a particular deficiency? Understanding this will help me decide which sort of program is most appealing to me – not a fully scientific/objective decision, since my view of the “good life” affects which disorders I most want to address.
  • In what region do the people most affected live? What age are they? Any special circumstances (i.e., pregnant women)? Before focusing on helping a group of people, I think it’s important to know what other obstacles they face. Personally, I want to help people for which malnutrition is the main (or a significant) obstacle to living a full, happy life.

  Next time, I’m going to try to find some answers as well as an organization to support.