Category Archives: BRAC

Giving Well: What Should Count Besides the Numbers?

Posted on by
Reply
5247123470_70285fefa2_b

Pictured: INEFFICIENT RESOURCE USE (Photo: “Citizens of Abyei Protest Bashir’s Statement” by ENOUGH Project)

It’s pretty common for sites and such devoted to helping you “give better” to ultimately look for some criterion of efficiency. For instance, the first two criteria that GiveWell uses for assessing charities are “strong evidence of impact” and “highly cost-effective.” Similarly, when Giving what we can declares that some charities are 1000 times more effective than others, they focus almost entirely on cost effectiveness. This leads them into some knots on their assessment of some kinds of interventions – for instance, they start by trying to translate things like education into economic improvement (can we measure how many $ of future income $1 of education spending creates?) and political advocacy into health spending (if $1 of spending advocating for bed nets gets the government to spend $3 on bed nets, it may be worth it). I think it’s unsurprising that trying to force things this way leads them to largely throw up their hands on things like education, political advocacy, and emergency relief, though that might be a conversation for another time. But, for instance, I think they’re actually too easy on emergency aid, by their own lights. Emergencies, by definition, are complex and chaotic environments, and so it’s going to be almost impossible for the cost of a disability-adjusted life year (DALY) in an emergency to be lower than the cost of a DALY in a situation where people are simply poor. So long as there is space for additional funding of things like anti-malarials, by the logic of cost-effectiveness, it will almost never make sense to go with, say, Doctors Without Borders over the Against Malaria Foundation.

I think this focus on a narrowly defined notion of efficiency is problematic. I’m not going to give a knock-down argument in a blog post for that, but I’d like to at least make some notes toward broadening the conversation.

The easy way to do this would be to say, “let’s just not focus on outcomes so much.” If I simply have an obligation to help some people and not others, regardless of how effective my help is, that of course makes this whole approach wrong headed. And, in fact, I tend to think that our obligations to help others might look more like Kantian imperfect duties than utilitarian maximizing, but it’d be dirty pool to start with that. I think there are things even the consequentialists should be keeping in mind.

And also, nothing I say here should be construed as saying that efficiency doesn’t matter at all. It’s certainly a worthwhile consideration. Even if you bought an almost wholly deontological picture of your obligations, if organization A accomplished the exact same goals as organization B, but at half the price, you should probably go with A.

Finally, I don’t think the upshot of these considerations is kumbaya-we-should-all-give-wherever-we-please. I do think it’s weird that people I know spend much more time worrying about the marginal efficiency of the charities they give to than they do to the question of whether the marginal dollar of their income should go to charity or personal consumption. It’s a tough question whether I’m doing more “good” by giving $1 to BRAC than by giving it to Oxfam, but it’s almost certain that giving that dollar to either does more good than spending it on a flavor shot in my coffee.*

It’s hard to measure politics.

It’s currently somewhat trendy to measure political advocacy in terms of leveraging your funds, especially since corporations get so much bang for their buck out of lobbying. And that’s something worth thinking about (though I wish people who make this argument would pay more attention to where the money is coming from – e.g., does it make sense to pay for bed nets out of Ghanaian taxes rather than out of my USian pocket?).

But to think that that’s the whole of it impoverishes our notion of politics. Even if you leave aside the inherent value of political participation, the changes that political change can wreak go far deeper than just the number of bed nets provided. I mean, Marx imagined the communist revolution as increasing the material luxuries available to the proletariat, but if you try to reduce it just to lowering the Gini coefficient of a society, you’re missing the point a bit. Politics isn’t just a way of setting spending priorities, it infuses people’s whole lives.

Lest you think this is an airy philosopher’s concern, keep in mind that there are important policy questions that turn on whether or not we think of the value of political systems primarily in terms of material benefits to the population. The whole history of supporting authoritarian governments on the theory that they can make the hard choices needed for economic growth is based on identifying the two.

Even if you do focus just on material change, the important changes may be missed if you focus just on the kind of thing we have a reasonable hope of measuring in a straightforward fashion. Sen has famously argued that democracies don’t have famines – this is a straight-up material benefit to democracy, but it would be at least fiendishly difficult to use that result to calculate the return on investment for each pound sterling contributed to Gandhi’s IndieGoGo campaign. More recently, Peter Buffett posted a somewhat-fluffy-but-makes-a-good-point column arguing that it’s weird to see the same captains of industry and government who create problems like poverty getting together to decide how to spend money to alleviate it. In other words, even if we focus only on material benefits, there’s the question of how many bed nets $1 of spending on lobbying for more bed nets buys, and then there’s the question of how we got to the place where people can’t afford bed nets in the first place.

There’s more to live than living.

I blame the Rawlsian doctrine of primary goods for this one. What about the inherent goods of education and democratic participation? Why should we care less about them than DALYs and the like?

The quick answer is that, if you’re not alive, you can’t enjoy anything else. But as I’ve argued more formally elsewhere, this is a misleading way to look at things. In a nutshell, if we really put an absolute priority on preserving our own lives, we’d all hang out in underground bunkers wrapped in bubble wrap (if you want the longer version, I have a chapter on it in my vaporware book, or you can just go read Butler and look cooler). We accept risks (which in the language of decision theory, just are reductions to our expected number of DALYs) all the time for things we think are important. The argument that one DALY is more important than, say, a child having access to the means to create art is one that we should be having, not one that we should assume away at the start (I mean, it’s a bit weird to me that almost no discussion of charitable giving seems to even notice the discussion about capabilities in the theory of development).

Issues like the symbolism of giving might live here, too. For instance, if I give to a poverty-alleviation organization based in the global South, that might have a good element of “saying” that I don’t think poor people are just victims whose problems need to be solved by white people like me. It might also have non-symbolic benefits mediated by the symbolism, like building organizational capacity in poor communities (often a criticism of aid agencies that swoop in with ready made programs, even if in the short run those may look more “efficient”) – this overlaps with the importance of politics, above – or just changing a hegemonic mindset that can tend to demoralize poorer people.

There’s a difference between badness and injustice.

One intuitively “punchy” reason to be attracted to groups like MSF that work in war zones and the like is precisely the thing that makes it hard for them to compete on pure efficiency measures: they work in war zones and the like.

This starts to push against the assumption that we’re playing nice with the consequentialists, but I don’t think it entirely breaks the rules. It’s at least plausible that there is something worse about someone dying because they got caught in the crossfire of a war than because they happened to catch malaria.

Now, this may pull against some of the considerations about politics, above. If most poverty is injustice rather than bad luck, the gap between the person who dies from a gunshot and the person who dies from poverty-induced malnutrition narrows. But again, I don’t think we get to ignore the conversation (and if you buy the equivalency, you probably should be working for structural political change more than most discussions of charity imply, since the injustice of poverty remains even if you mitigate its effects by giving someone a bed net. If you steal my money, and then someone gives me the P.O.S album I was going to buy with it, it doesn’t morally sanitize the theft).

We might have other special obligations.

This is dicey territory. The push towards efficiency as the overriding moral criterion for giving comes from the – quite powerful – idea that I should measure my action only by how much good it does, and not by morally arbitrary criteria like how close I happen to live to the recipients of my assistance.

But this flattens out plausibly important distinctions. Again, this bends the rule of being nice to consequentialists, but at least pluralist consequentialists can take it seriously.

I might be required to put my thumb on the scale for people whose poor circumstances are in part my fault. For instance, it seems plausible that, as an American, I might have a greater responsibility to help out victims of the wars in Iraq and DR Congo, where my country had a strong hand, than I do to people in some place like (uh… shit, the US has its fingers almost everywhere…) maybe Mali, which is more on the French’s moral account.

I might have special obligations to do good in my own community. For instance, I spend a lot of time these days on prison teaching and (increasingly) on getting involved with violence-reduction in my home city. That’s time that I could be getting a second job and donating the money to buy bed nets.

Now, the worry here is that allowing this kind of consideration opens the door to all sorts of moral abuse. It’s a very compelling worry that at some point, when I’m giving money to help out with re-greening the golf course in my gated community** I’ve crossed the line into just buying luxuries for myself (if I golf) or being morally self-indulgent (if I don’t) under the guise of “charity.” I agree that there needs to be some line here!

But I think it points to the fact that the distinction between consumption and altruism that seems taken for granted by the conversation about giving is too blunt an analytical instrument. It seems right to question whether the prison teaching work does as much good as donating the salary I could earn in that time to the Anti-Malaria Foundation.*** At the same time, it seems odd to count it as just a fancy kind of self-interested “consumption.”****

One utilitarian way around the weirdness here would be to instrumentalize it. Dollar for dollar, my investment in prison teaching is inefficient and sub-optimal. But human nature being what it is, I am more likely to sustain altruistic activities with a face-to-face component, so it makes sense to do this as a way to maximize the good I do in the teeth of human weakness of will. Basically, I’m much more likely to quit the second job where I send money to a faceless organization helping people buy bed nets than I am to quit the work where people tell me directly that I’ve helped them. It’s not right, it’s just a prediction about my own psychology, which is just another utilitarian datum.

But this strikes me as not quite right. I think we need to de-impoverish our conceptual apparatus for thinking about altruism more. I’m not sure we need to “flatten” things like working in one’s own community into either a category of self-interest (where we make a space for them alongside other forms of permissible consumption) or charity (where they probably compete poorly with bed nets). They can be their own category, at least if we’re willing to go at least as far as value-pluralist consequentialism.

And the moral motivations are at least prima facie different. When I give to BRAC, I do it because I think as a relatively affluent person (I AM THE 17%) I am obliged to help out those less fortunate, in a very generic sense (I also think BRAC scores high on some of the less quantifiable stuff above, but I’ve already written about them). I do the prison teaching because I feel like I have a special obligation to make my own community a better community. It’s not clear why these can’t coexist, or even need to be ranked.

Fuck, maybe it is just Kantian imperfect duties. Sorry, utilitarians.

* Blech.

** Fuck gated communities. They’re necrospaces.

*** This also opens the door to the question of why I work as an academic when I could have gotten a higher-paying job as a lawyer or businessman if I’d followed a different path.

**** I mean, yes, inevitable “there’s no such thing as altruism” assholes, I get warm fuzzies from the work. But I get warm fuzzies from donating to BRAC, too. Go read your Hume, but also that argument doesn’t make the distinction here at least.

Giving Well: Oxfam versus BRAC

Posted on by
"A Hindu Woman Giving Alms," by Raja Ravi Varma

“A Hindu Woman Giving Alms,” by Raja Ravi Varma

Daniel Levine has an interesting discussion of giving and giving well up today on whyiamwrongabouteverything:

When I got a “real” job at USIP, back in 2007, I resolved that I was going to donate 10 percent of the portion of my take-home pay that I kept for personal use (as opposed to what I contribute to the joint account I share with my wife). This is less than the Giving What We Can pledge, but more than the The Life You Can Save pledge, so I figure it’s at least a good start. (My wife and I also give 5% of the after-tax income we contribute to our joint account).

Some may think it impolite, but I actually really appreciate that Daniel laid out his giving budget. 5% of joint household funds and 10% of personal funds dominates my giving budget quite a bit: last year we gave about 3% of our total pre-tax income to Oxfam (which is similarly ignored by GiveWell) so his commitments and reasons are particularly impressive. My family will likely scale back this year to make room in the budget for my wife’s unpaid maternity leave, but now social competition will give us an incentive to increase it!

He also highlights his preferred charity, the Bangladesh Rehabilitation Assistance Commission. Despite the name, BRAC actually works in eleven different countries in Asia, Africa, and the Americas. Because they are primarily funded by micro-finance, they are able to fund 80% of their own charity work. Donors pay for the rest. Yet the charity rating service GiveWell refuses to rate it. Before I met Daniel, I’d never heard of BRAC, and since then I’ve been trying to do my homework. Two potential objections have emerged:

  • Microfinance often involves usurious interest rates. Should we worry that this charity is largely funded by some of the people who most need charity? Hugh Sinclair reports that BRAC charges women in South Sudan 88% interest! If this isn’t as much of a problem as it seems, then why shouldn’t we see the self-funding model as sufficient, and direct our donations to organizations that cannot self-fund?
  • Is there any reason beyond symbolism to prefer a South-South charity to a North-South charity? And is the symbolism worth potential inefficiencies or less-than-optimal life-saving?

All of my objections boil down to a simple concern: is BRAC better than Oxfam? Perhaps Wrongzo’s nom de plume is a misnomer or humblebrag, but perhaps he and I have a disagreement, after all.

Academics like to distinguish between two questions: whether we can know the right answer, and whether there can *be* a right answer. William Easterly rejected this kind of mythology of metrics when he told Peter Singer that “it is not at all clear that you (or anyone else) knows exactly what to do to save the lives of poor children or how to get them out of extreme poverty.”Perhaps Easterly is right to answer the first question with skepticism, but I believe we can answer the second question affirmatively: there is a “best” use for my money, some single expenditure that reduces suffering the most, even if we do not know what it is. People who criticize measurement efforts effectively admit that there’s such a thing as a right answer to the question of how to give, because they believe that Peter Singer and the metricians have the wrong answer!

If there’s a right answer, why not take a shot at figuring it out? So: is Wyclef Jean’s charity effective? A little research suggests not. The key here is that there may be many good answers, but there are certainly some bad answers, too. For instance: donating money to a church or to a museum doesn’t save lives, so those are demonstrably inferior kinds of charity. Even worse, sometimes our helpful efforts are actually harmful, as when we learned that some arsenic mitigation efforts may actually increase infant mortality! Whether microfinance at 88% interest is good or bad for a community is a matter that can be evaluated independently. And what’s more, it may be helpful even though it seems, to me, to be exploitative!

Yet it’s also possible that this is such a really hard problem that we’re better off with no information rather than some information. In donating to large charities with expansive internal research arms, we are essentially using part of our dollar to buy evidence about where the rest of the dollar should go. This seems wasteful! At some point, you reach diminishing returns in terms of the evidence-costs versus the marginal utility gains. Perhaps there really is no reason to believe that Oxfam’s or GiveWell’s internal metrics will help me direct my money better than traveling to a poor country and handing out twenty dollar bills! You may think I’m joking, yet I’ve actually seen this proposed by the economist Tyler Cowen. Perhaps we should skip airfare [overhead!] and simply mail the money to a random person!

That’s one reason I prefer larger institutional charities as informal indexers, in the portfolio sense: administrative/information costs are higher, but smaller as a percentage of total expenditures, while new money is always redirected to the best-informed current needs. Today’s best charities may not be tomorrow’s best charities. We know that, for instance, only about 50% of the population needs to be mosquito-netted to get almost all of the health effects. So at that point, it’s time to redirect the resources to a new cause, from malaria to diarrhea, say, or else new dollars could have no utility at all.

If we’re constantly analyzing the productivity of a charity, like GiveWell and Oxfam do, we’re likely to catch it. But if we’re sitting back and receiving the reassuring development letters from the charities’ staff, we’re likely to irrationally remain committed to the “less-than-best charity” for long after our donations have stopped having the optimal utility. That’s something Oxfam can do but VillageReach can’t. From the research I’ve done, I can’t tell whether BRAC is doing so or not; this was GiveWell’s problem in 2009 when they last evaluated BRAC.

You may well wonder: why even argue about charity? Shouldn’t we just give quietly and privately?

The various academics associated with Giving What We Can are engaging in a conscious effort to change the norms and standards of charitable giving. It’s true that donors mostly give for reasons of self-satisfaction, which is why consequentialists of various stripes are engaged in a quiet effort to change the conditions under which donors can successfully congratulate themselves. By working on the codes of honor and merit, they hope to have an outsized impact on the behavior of major givers and institutions. Academics recognize that we’re not rich and powerful, but we like to think that words and arguments can sometimes give us a bit of a multiplier effect.

To some extent, they’ve already succeeded, such that you see major criticisms of goals within global health and humanitarian aid communities for ineffective models, like the work of William Easterly, Dambisa Moyo, and David Rieff. More recently, some in the aid community have questioned the cost-benefit efficacy of the Gates Foundations’ attempts to eradicate polio.

But this requires a pretty strict consequentialism (though not utilitarianism) to which many retail donors object. Beyond the overall skepticism about knowledge and metrics, there’s an underlying fear that consequentialism levels the playing field between giving and consuming, and that this will become far too demanding for the average donor. Once you get started down this path, giving well goes from an analytic tool to a duty. It starts to sound overly demanding, and maybe even a bit melodramatic, like we’re all in the same position as Oskar Schindler:

This car. Goeth would have bought this car. Why did I keep the car? Ten people right there. Ten people. Ten more people. This pin. Two people. This is gold. Two more people. He would have given me two for it, at least one. One more person. A person, Stern. For this. I could have gotten one more person… and I didn’t! And I… I didn’t!

In the film, you’re supposed to think he’s being too hard on himself. But isn’t he right? Ten people died so some rich industrialist could drive around in luxury. How many died so that I could sit up late typing this post on my computer? How can that possibly be just?

It’s a common intuition: whenever we see a rich person spending lavishly on a boat or a sports car, don’t we sort of feel that they’re wasting their money, that there are folks in need who could use it better? Am I really a philistine for not appreciating the craftsmanship in a Porsche or the softness of 600-count Egyptian cotton? I like lots of luxury items, too: I’m not an ascetic. Right now, I’m lusting after the new suit, an expensive rowing machine, and lots of electronics that are totally unnecessary. I may even buy some of that stuff. But I think we should admit that it’s not particularly praiseworthy to spend my money on luxury goods while there are children dying from diarrhea and women living with obstetric fistulas. We could treat an obstetric fistula for $450 dollars. That’s less than an iPad! It seems like an easy choice, yet I’ve already spent more time dithering on the minute differences between BRAC and Oxfam than I do wondering whether to spend the next $450 I make on consumption or charity.

Why I’m Giving Money to BRAC

Posted on by
Reply
168753_10150122106404680_507169679_7931812_4412112_n

Lauren and me at the final “fancy pants” party, summer 2006. Photo by Lindsay Moore.

Hey, for anyone expecting another combative post, sorry! This is going to be a bit more personal and boosterish.

On 22 January 2011, my friend Lauren Fleming died. I’ve already said about everything I have to say about it to about everyone I know who knew her, so this isn’t really going to be a post about her.  It’s just a bit of context.

More context, but less important. When I got a “real” job at USIP, back in 2007, I resolved that I was going to donate 10 percent of the portion of my take-home pay that I kept for personal use (as opposed to what I contribute to the joint account I share with my wife). This is less than the Giving What We Can pledge, but more than the The Life You Can Save pledge, so I figure it’s at least a good start. (My wife and I also give 5% of the after-tax income we contribute to our joint account).

I used to make my charitable donation as a lump sum on or around my birthday (since charitable organizations incur transaction costs on donations, other things equal you’re better off donating in larger amounts to a smaller number of organizations). But since Lauren’s death, I’ve been donating on or about January 22, in memoriam.

"Loan Process" by IFPRI-IMAGES on Flickr.

“Loan Process” by IFPRI-IMAGES on Flickr.

This is really just context and prologue to what I really wanted to talk about here, though, which is the organization I’ll be making my donation to this year: BRAC.

BRAC was founded as the Bangladesh Rehabilitation Assistance Commission (though the acronym seems to have floated free of that particular meaning).  Apparently it was originally just their founder putting some persons displaced by the cyclone up in his offices at Shell, where he was an accountant.

They have since expanded their operations massively, becoming by some reports the largest NGO in the world, with operations well beyond Bangladesh – it is also believed to be the largest NGO operating in Afghanistan (for most of the empirical claims, see the linked Economist and SSIR articles, by the way), and has operations in Pakistan, the Philippines, Sri Lanka, Liberia, Sierra Leone, South Sudan, Tanzania, Uganda, and Haiti.

A big part of BRAC’s core operation is microfinance, but they follow a “microfinance-plus” approach.  That is, one of its core businesses is microfinance, but it also provides education services (including a university), some advocacy, community-building, health care, etc.

There are a few key reasons to like BRAC.

First, and in many ways foremost, “it is a [global] Southern organisation from a poor country expanding to other poor Southern countries.” (10)  This is important for moral and ideological reasons. It breaks up invidious tendencies to cast poor countries always in the victim, never in the agent role. It has less of a neo-colonial savor, which isn’t nothing. On a more direct, but still moral, level, it means giving money to aid workers who do not enter their work with the social and economic baggage that many Western aid workers do.  For instance, BRAC employees are paid wages that are high for locals, and for Bangladeshis, but much more in line with local wages and standards of living.  The Southern origin of BRAC is also practically important – it means that many of its workers will have a deeper understanding of the global poor live than even the best-informed and sincerest Westerner.

Second, it ranks high on sustainability. Though BRAC does accept donations (hence this post), about 80% of their operating budget comes from their own activities (like the microfinance and seed banks).  Thought I’d throw something in there for my more business-minded readers.

Third, BRAC is very much involved in “frontline” services. They have a small advocacy unit, but the bulk of their work is still service provision, in a time when lots of the big Western NGOs are moving towards policy influence. Your mileage may vary on this – I have some skepticism about the ability to influence policy from the top-down, plus it’s my day job (sort of), so when I give money to aid I like to give it to groups that are largely focused on service provision.

There’s a worry here that I may have more to say about later on this blog – doesn’t direct service provision create dependence or a parallel government (that gets the real government off the hook)?  I worry more about the latter than the former – I don’t really worry about the former at all – but at the same time aid conditionality has had a mixed track record at best. Where governments are bad, I don’t see a lot of evidence that starving them of aid makes them much better (not that we shouldn’t care about making them better), especially when the aid is direct service provision rather than support to the government’s own coffers.  And direct support loses much of its moral lustre when the government already isn’t accountable to its people.

Fourth, on the other hand, warming the cockles of my policy-analysis heart, BRAC has a strong internal committment to monitoring and evaluation, including having a well-respected in-house research arm.  Even groups critical of it – more on that in a sec – note its committment to evaluation. In addition, BRAC is pretty heavily studied – and generally found to have a positive impact on poverty and development – by outside academics.

Fifth, BRAC has, since 2002, been doing innovative work to help the “ultra-poor.”  In a nutshell, the problem is that people at the very bottom of the socio-economic ladder are often unable to benefit from the traditional development programs to get a leg up – e.g., microfinance or access to education may not help them much because they are too ill and malnourished, or too socially marginalized, to make effective use of them.  To my eye, one of the most intriguing components of their approach is the community involvement angle. When BRAC starts up a poverty targeting program, one of the first steps is to engage the community in a participatory process of identifying the community members most in need and strategies to help them.  Community involvement in the resources and the progress of those poor is then  part of the ongoing strategy for addressing poverty. In part, apparently, this comes from the founder’s sensitivity to the ways in which social stratification in Bangladesh interfered with people coming out of poverty, which shows real insight (or maybe I’m just biased since it fits with the view I articulate in my own book, that social domination is the ur-vulnerability).

BRAC is, of course, not without its critics. In particular, I was a bit surprised and disconcerted to learn that GiveWell, an aid-evaluation org that lots of “let’s give more of our money to the poor” projects (like the two I link above) rely heavily on, does not recommend giving to BRAC.

But I will be honest that I cannot really tell why. The stated reason on their assessment page is that the research arm of BRAC does not provide any public evaluations of their employment or income programs (at the time of the 2009 review). But this seems to be at least partly inaccurate.  For instance, BRAC’s research arm has a report on the employment impact of their grants to small and medium enterprises, dated 2005 (GiveWell claims they searched on relevant terms, but I found it with “employment”).

In addition, GiveWell seems to move the goalposts in their assessments.  The only one of their top three charities that focuses on poverty alleviation (as opposed to medical research and implementation) is GiveDirectly, an org that transfers money directly to poor people in Kenya. Now, I have no problem with GiveDirectly – and I like their model, too. But GiveWell counts them as effective because there is evidence that they do in fact distribute their money, and there is no strong evidence (they feel) that the presumably good impact of giving poor people money is either not as good as it might seem, or that it is undercut by other bad effects. Meanwhile, their bar for BRAC is higher – not just that they conduct their programs, but that they be able to prove they alleviate poverty.

I’m happy to chat more about aid assessment methodology – it’s not my direct area of expertise, and something I’m still learning myself. But it leaves me pretty comfortable that giving to BRAC is a pretty good use of my money.

Of course, it doesn’t bring Lauren back.